ENVIRONMENTAL ANALYSIS
OF THE URANIUM FUEL CYCLE
PART III ^Nuclear Fuel Reprocessing
October 1973
U..S. ENVIRONMENTAL PROTECTION AGENCY
Criteria and Standards Division
Washington, D.C. 20460
-------
KllVIKONHKIM'AI. ANALYSIS Of' Ti'K IIHAIIIUit FHKI, CYUi;
PART m - HU(;i,i',Ais .I'ur.i. KKPUiM'.KSsi)•«;
KKKATA
Paf.e 9., line 1H - "annual role.s" should road "max iimim annual, doses"
Pant: 9, line 20 - detet.e "both annual and"
P.if.R 9, line 2.0 t- "tloso.s" Hlnnild read "dose roimn.! t.ruenl:r>"
P«ip,(» .10, table 2, caption l:.it:.1c - "Poinilation DOKC" should re;id "'Population
Dose Commitment Inc.reinnnl.'s"
Pago 10, Lnhje 2, th.i rd column - add t:hc caption "Maximum Individual DOKC"
Page 30, table 2, fourth thru sixth columns - (he c.ncion should rend
"Average Orp,an Do.se Cloiniti.itme.ivt from an Annual Release (J9BO startup)"';
the units are "ninn-rein", not "mnn-vem/yr"
Page 10, table 2, fiftli 'column - "U.S." should read "rest of U.S."
9
Pago J.A, figure 1 - the. I)F for Pu-239 should read 10
Pages 15-18, figures 2, 3, and /i - the curves labeled "future" represent
the sum.of health effects induced by both future and past exposures.
Page C-16, line 23 -.add after "individuals", "in a population of 10 "
Page CrH, line 2 - "r.ad" should read "30 rads" and add after "year", "at
a dose rate of one rad per year"
Mini' .,1 1.1 „ ) 'iin"n
Page C-21, line 8 - "10" sliould read "10
Page D-17, last line - t» should read t1
Page D-22 - PW = ACXpwf
Page D-23 - Correct last column to reflect above
Page 22 - Correct 3rd and 5th columns to reflect above,
Figurcr. 6, 7, l>-2, l)-3 - Correct to reflect above.
-------
FOREWORD
The generation of electricity by light-water-cooled nuclear power
reactors using enriched uranium for fuel is experiencing rapid growth in
the United States. This increase in nuclear power reactors will require
similar growth in the other activities that must exist to support these
reactors. These activities, the sum total of which comprises the uranium
fuel cycle, can be conveniently separated into three parts: 1) the
operations of milling, conversion, enrichment, fuel fabrication and
transportation that convert mined uranium ore into reactor fuel, 2) the
light-water-cooled reactor that burns this fuel, and 3) the reprocessing
of spent fuel after it leaves the reactor.
This report is one part of a three-part analysis of the impact of
the various operations within the uranium fuel cycle. The complete
analysis comprises three reports: The Fuel Supply (Part I), Light-Water
Reactors (Part II), and Fuel Reprocessing (Part III). High-level waste
disposal operations have not been included in this analysis since these
have no planned discharges to the environment. Similarly, accidents,
although of potential environmental risk significance, have also not been
included. Other fuel cycles such as plutonium recycle, plutonium, and
thorium have been excluded. Insofar as uranium may be used in high-
temperature gas-cooled reactors, this use has also been excluded.
The principal purposes of the analysis are to project what effects
the total uranium fuel cycle may have on public health and to indicate
where, when, and how standards limiting environmental releases could be
effectively applied to mitigate these effects. The growth of nuclear
energy has been managed so that environmental contamination is minimal
at the present time; however, the projected growth of this industry and
its anticipated releases of radioactivity to the environment warrant a
careful examination of potential health effects. Considerable emphasis
has been placed on the long-term health consequences of radioactivity
releases from the various operations, especially in terms of expected
persistence in the environment and for any regional, national or world-
wide migration that may occur. It is believed that these perspectives
are important in judging the potential impact of radiation-related
activities and should be used in public policy decisions for their
control.
Comments on this analysis would be appreciated. These should be
sent to the Director, Criteria and Standards Division of the Office
of Radiation Programs,
W. D. Rowe,
Deputy Assistant Administrator
for Radiation Programs
lii
-------
-------
CONTENTS
Page
FOREWORD. iii
INTRODUCTION 1
EFFLUENT CONTROL TECHNOLOGY 6
RADIOLOGICAL IMPACT OF A REPRESENTATIVE PLANT 8
CUMULATIVE RADIOLOGICAL IMPACT OF THE INDUSTRY 13
ECONOMICS OF HEALTH EFFECTS REDUCTION AND EFFLUENT CONTROL... 20
SUMMARY AND CONCLUSIONS. 25
REFERENCES 29
APPENDIXES
A. Spent Nuclear Fuel Radioactivity Forecasts
B. Fuel Reprocessing
C. Radiological Impact of Fuel Reprocessing
D. Economic Impact Analysis
-------
FIGURES
1. Estimated Cumulative Environmental Buildup of Radioisotopes from the
Total Fuel Reprocessing Industry in the United States. ....... ....... 14
2. Estimated Past and Future Health Effects Committed by Tritium
Releases from the United States Fuel Reprocessing Industry ........ , . 15
3. Estimated Past and Future Health Effects Committed by Krypton-85
Releases from the United States Fuel Reprocessing Industry ...... .... 16
4. Estimated Past and Future Health Effects Committed by Iodine-129
Releases from the United States Fuel Reprocessing Industry .......... 17
5. Estimated Past and Future Health Effects Committed by Actinide
Releases from the United States Fuel Reprocessing Industry ..... ..... 18
6. Risk Reduction vs Control System Cost . ........ ............... ..... , . 23
7. Control System Cost Effectiveness vs Control System Cost ....'. .. ..... 24
A.I Estimated Cumulative Environmental Buildup of Tritium from
the Fuel Reprocessing Industry in the United States . . « ......... A-ll
A. 2 Estimated Cumulative Environmental Buildup of Krypton-85
from the Fuel Reprocessing Industry in the United States. . ..... A-12
A. 3 Estimated Cumulative Environmental Buildup of Iodine-129
from the Fuel Reprocessing Industry in the United States ....... A-13
A. 4 Estimated Cumulative Environmental Buildup of Plutonium-239
from the Fuel Reprocessing Industry in the United States ....... A- 14
B . 1 Typical Process Flow Schematic ......... ........ ........... ..... B-4
B.2 Schematic Diagram Showing the Steps Required for the "Zero
Release" Reprocessing Concept .................................. B-ll
D.I Projected Total United States Population ..... ............ . ..... D-7
D.2 Risk Reduction (Mortality and Morbidity) vs Cumulative Control
System Cost . . ............ .... ............ ........ ...... ........ D-25
D.3 Risk Reduction (Mortality) vs Cumulative Control System Cost ... D-26
vi
-------
. TABLES . PAGE
1. Site Characteristics of Nuclear Fuel'Reprocessing Plants 5
2. Projections of Average Annual Population Dose from-a 5 tonne/day
Nuclear Fuel Reprocessing Plant ,.,, 10
3. Projections of Total Health Impact from a 5 tonne/day Nuclear
Fuel Reprocessing Plant. 12
4. Summary of Health Effects and Costs of Emission Controls for a
5 tonne/yr Fuel Reprocessing Plant 22
A.I Estimated U.S. Fuel Reprocessing Requirements A-2
A.2 Representative Quantities of Potentially Significant Fission
Products in Spent Reactor Fuels A-6
A.3 Representative Quantities of Potentially Significant
Activation Products in Spent Reactor Fuels A-7
A.4 Representative Quantities of Actinides Present in Spent
Reactor Fuels A-9
A.5 Estimated Annual Inventories of Selected Nuclides in Spent
Reactor Fuels-........'. • A-10
B.I General Information for Commercial U.S. Nuclear Fuel
Reprocessing Plants B-2
B.2 Control System Data for Nuclear Fuel Reprocessing (LWR +
Recycle Fuels) . • B-15
Q e
B.3 Comparison of Processes for the' Removal of Kr from
Dissolver Off-Gas from a Fuel Reprocessing Plant B-17
C.I Milk Concentrations of Iodine-131 and Iodine-129 from Given
Input 'Concentration and Corresponding Doses ...,,....,.,... C-9
C.2 Actinide Air-Dose Conversion Factors Relative to. Plutonium-239«» C-ll
C. 3 Summary of Media-Dose Conversion Factors • • •« C-12
D.I Factors Used in -the Assessment of Environmental Radiological
Impact of Fuel Reprocessing ...,., D-19
D.2 Estimate of the Economics of a 5 MTU/day Reprocessing Plant D-21
D. 3 Control System Costs • D-23
vii
-------
-------
PART III - NUCLEAR FUEL REPROCESSING
IMTRODUCTIOM
Economic analyses performed both by the AEC and by commercial
investors have concluded that the economical generation of electric
power by nuclear plants requires that valuable isotopes of uranium and
Plutonium be recovered from spent reactor fuels for re-use in new fuel
elements. Recovery of uranium and plutonium involves mechanical
chopping of spent fuel elements into small pieces and placing them in an
acid dissolver to separate the spent fuel from its metal cladding prior
to chemical separation of useful Isotopes from waste products by some
adaptation of the Purex solvent extraction process. This operation
results in the controlled release of fission products and other
radioactive waste materials which accumulate in the elements during
burnup in the power reactor. Since the quantities of these waste
materials are large, considerable care is taken to assure that the fuel
elements maintain integrity through the cycle in the reactor. In
essence, therefore, the fuel reprocessing step breaks this carefully
constructed barrier and, as a consequence, represents the main source of
radioactivity from the nuclear power industry which could potentially
enter the environment.
Many complex technological, environmental, and biological factors
are involved in judging the impact of radioactivity on the environment.
It is the purpose of this analysis to examine these factors with respect
to fuel reprocessing requirements over the next several decades in terms
of the potential public health and environmental risks involved.
-------
The main objectives of this review are twofold; (1) to determine the
population health risks and the-cost effectiveness of waste controls
relating to these risks for the present and future auclear fuel
reprocessing industry, and (2) to document the data base and techniques
for assessing the environmental,impact of the fuel reprocessing industry.
A number of•considerations are involved in accomplishing these objectives.
These include forecasts of fuel' reprocessing requirements through the
year 2020, a detailed analysis of effluent control systems, the environ-
mental transport of radionuclides, and considerations of the resultant
doses and health effects. The data base and techniques used for these
-analyses are presented in detail in the appendixes of this report.
These studies are based on the performance of a hypothetical
reprocessing plant having characteristics typical of plants now under
construction. This model plant was used to evaluate potential
reductions in health effects for various controls that limit the release
-of radionuclides to the environment. Control system costs were
considered as well as .their relation to. total plant costs. These
analyse's were then expanded to project the impact of all the nuclear
fuel reprocessing operations expected in the United States up to the
year 2020. , Doses and world wide health effects have been estimated on
the basis of projected nuclear power production and anticipated
radioactive effluent control .techniques. Only the contribution from
United States reprocessing facilities and only the most significant
radioisotopes, in terms of total effect and persistence in the
-------
environment, have been considered in this analysis. Major changes in
normal operating procedures, such as shorter cooling time (storage
between fuel discharge from the reactor and initiation of reprocessing)
should significantly affect local and regional doses but would not
appreciably affect national and world impact. >
• . •Three reprocessing plants, with an estimated combined annual
capacity of 2,700 metricjtons of_ju^anium_QMTU), are expected to be in
operation in the United States in the next several years. The fuel
reprocessing industry is expected to increase along with nuclear power
growth to about 50 to 60 plants with reprocessing capacity of about
80,000 Mill per year projected by the year 2020,
Up to the present, only one commercial facility has been
operational, but is now shut down for modification. This is the tfuclear
Fuel Services plant in West Valley, New York, with a capacity of 1 tonne
of uranium per day. Processing of spent nuclear fuel was initiated in
1966 and continued with sporadic interruptions until December 1971 when
the plant was shut down to permit expansion of processing capability
which will increase plant capacity to 3 tonnes per day. The plant is
located on a 3,300 acre tract owned by the State of New York in Ashford
Township, Cattaraugus County, New York (Ref. 1). Buffalo, New York, is
26 miles from the plant and several of its southern suburbs are within a
25 mile radius. Considerable dairy farming and other agricultural
activities are conducted close to the site. New York State conducts a
comprehensive monitoring program around the facility including daily raw
milk surveillance (Ref, 2 and 3). Extensive studies have been conducted
-------
at th±s facility by EPA and its predecessor organizations (Ref. 4, 5, 6,
and 7).
Two other plants are under construction. The Midwest Fuel Recovery
Plant is a tonne per day plant located at Morris, Illinois, adjacent
to the Dresden Nuclear Power Station on privately owned property (Ref.
8). The General Electric Company, which has received a permit to store
spent fuel and process unirradiated fuel, owns and operates the plant,
Opefations with irradiated fuel from nuclear power reactors will
probably be started during the summer or fall of 1973.
The closest population center is Joliet, Illinois, 14 miles from the
plant. Part of the city of Chicago is included within a 50-mile radius
of the plant, thus significantly increasing the local population of
concern. Offsite concentrations must be evaluated in terms:of a
multiple source since the three Dresden reactors are in close proximity.
The Barnwell Nuclear Fuel Plant is designed to reprocess 5 tonnes of
uranium fuel per day (1,500 tonne per year), and is located adjacent to
the Savannah River Laboratories (SRL) in Barnwell County, South
Carolina, on privately owned property (Ref. 9). Construction of this
facility was begun during the spring of 1971, and the projected
completion date is 1975.
Augusta, Georgia, which- is 31 miles from Barnwell, is the nearest
population "center. The. population within 50 miles of the facility is
about 500,000 or a factor of 10 below the corresponding population at
Midwest. Additional site characteristics of these three facilities are
presented in table 1.
-------
TABLE 1
Site Characteristics of Nuclear Fuel Reprocessing Plants
'
Capacity
MTU/ day
MTU/year
Population within
10 miles
25 miles
50 miles
Population Center
(10 CFR 100)
Maximum Off-site
Annual Average x/Q
(sec/m^)
(1) Technical Specifications
(2) Adapted from ORNL-4451
NFS
1 (3)
300 (900)
21,409
293,711
Buffalo
(532,759)
26 miles
IxlO"6 (1)
(2.2xlO-7) (2)
for NFS (Ref. 1)
(Ref. 11)
MIDWEST
1
300
25,000
303,460
6,300,000
Joiiefc
(67,000)
14 miles
2.8xlO~8 (3)
(L.lxlO~7) (2)
BARNWELL
5
1,500.
7,100
65,209
60}, 084
Augusta
(70,600) -
31 miles
5.7xlO~8
•-
(3) Dresden stack use assumed
-------
6
The Western Interstate Nuclear Board has been assessing the need for
a fuel reprocessing facility in the Western region. It has concluded
that such a facility will be required in the late 1970's but may be
profitable during the middle 1970's (Eef. 10).
EFFLUENT CONTROL TECHNOLOGY
Current practices and effluent control measures in the nuclear fuel
reprocessing industry are not totally indicative of the workload and
performance to be expected in new plants coining online during the period
discussed in this report (1970-2020). Because newer designs will differ
substantially from the first commercial facility, Nuclear Fuel Services?
presently available performance data are of limited use in predicting
environmental releases. The two newer plants, the Midwest Fuel Recovery
Plant at Morris, Illinois, and the Barnwell Nuclear Fuel Plant at
Barnwell, South Carolina, are far enough along in their design and
construction to provide some idea of what the cost and expected
performance characteristics of new plants may be. Both of these plants
will utilize the recycling of liquid wastes so that in theory the only
radioactive liquid discharges will be from spills, accidents, or leaks.
The tritium present in the fuel will be released as a stack gas. This
approach is in contrast to the NFS facility which uses a system designed
to release low-level radioactive liquid wastes into the environment.
Waste control systems under construction for use in fuel
reprocessing plants and their costs are described in appendix B. The
design of NFS and the two plants under construction calls for the
release of radioactive krypton-85 as stack gas, although the owners of
-------
Midwest and Barnwell have indicated a willingness to add krypton removal
equipment when it is considered practical (Ref. B-38 and B-39). Removal
efficiencies of at least 99% are expected for such systems, although an
efficiency of 99.9% has been claimed by some vendors (Ref. 12). Other
effluent control systems to be incorporated in these plants will include
iodine removal systems, which are expected to retain more than 99.9% of
the iodine-129 and iodine-131, and particulate filters which should
3
reduce releases of the actinides (particulate forms) by a factor of 10
greater than the decontamination which is effected by the chemical
separation process. While the operating experience at NFS confirms the
effectiveness of particulate filters (Ref B-41), the anticipated
efficiency of silver zeolite which provides backup to the aqueous
scrubber iodine removal systems has not been verified in operational
situations. The experience at HIS with a different but modern iodine
recovery system has not been as good as orginally predicted (Ref. B-40);
this equipment is now being replaced with a silver zeolite system. The
radionuclide releases used in this report are estimates based on current
knowledge of the capability of waste control technology. Technological
innovation in waste control systems may reduce projected releases, while
maintenance problems in aging equipment may result in increased
environmental contamination.
-------
RADIOLOGICAL IMPACT OF A REPESENTATIVE PLANT
The assessment of the local, regional, or worldwide radiological
health Impact that may result from operation of a particular^ source or
distribution of sources of radioactive pollutants is dependent upon a
number of assumptions. The calculational models used are generally
categorized as follows:
1. Source term models.
2. Environmental transport models.
3. Dose models
4, Risk models.
Source term modeling includes assessments of pollutant generation
rates, inventories, and physical and chemical characteristics and
release rates to the hydrologic or atmospheric carrier.
The environmental transport model permits an estimate of the media
(air, water, food chains, etc.) concentrations at a particular point in
the space-time continuum. These models considered all important
pathways from the source to the receiver.
The dose model allows an estimate of the energy deposition and biological
effectiveness in a biological system that results from exposure to the media
concentrations. It considers such variables as ingestion rates, in-vivo
distributions and biological half-life, and energy deposition in critical
organs.
The risk model provides for estimation of biological effect
resulting from doses due to ionizing radiation.
A useful approach to evaluating the potential environmental impact
of fuel reprocessing plants and related control costs is to consider a
single plant which is representative of current technology and design.
-------
Such a representative plant has been assumed to have a capacity .of 5
tonnes per day, or an annual capacity of 1,500 tonnes over an operating
lifetime of 40 years. The fuel mix to be processed was assumed to
consist of equal amounts of uranium and plutonium fuel. Of the residual
waste product and fissionable material in the fuel, only the following
specific radionuclides are considered in this analysis: krypton-85, .
tritium, iodine-129, iodine-131, plutonium-239, and other actinides.
The release of other nuclides into the environment is anticipated to be
Q
less than one part in 10 and will not produce health effects comparable
to those produced by the nuclides considered here (Ref. D-8), although
the situation for ruthenium is unclear at present.
Projected amounts of the radionuclides in spent fuel are given in
table A.2 through A.4, appendix A. Liquid radioactive releases are
assumed to be insignificant, and all environmental releases are assumed
to be via the air pathway. The population within 80 km of the model
plant was estimated to be 1.5 x 10 people by the year 1980 and was
projected to double during a 40-year plant operating life (see appendix
D). Using these assumptions, annual dose rates from the various
nuclides were calculated (appendix D) for individuals located 3
kilometers from the plant, and both annual and total population doses,
measured in person-rem, were determined for: (1) the population within
80 km of the plant, (2) the total United States population excluding
those persons residing within the 80 km zone, and (3) the world
population excluding the United States. These doses are listed in table
2. For tritium, most of the United States population dose results from
-------
10
TABLE 2 .
Projections of Average Annual Population Dose
from a 5 Ml/Day Nuclear Fuel. Reprocessing Plant
Radionuclide
Source
Kr-85
H-3
1-129
1-131
Actinides
— Critical
Organ
Whole-body
Lung
Skin
Gonads
Whole-body
Gonads
Thyroid-
infant
Thyro id-
adult
Thyroid-
in£ant
Thyroid-
adult
Lung
Average
mreia/yr .
@ 3 km
0,38
0.75
13
0,50
3.2
3.2
1.4
.4
13
.8
1
Organ Dose from One
taan-retn/yr .
Regional
24
47
790
w
200
100
2.3
27
20
53
64
Year's Release for 1980 Startup
man-rem/yr . man-rem/yr .
U. S. World (less U.S.)
520 8,100
1,000 16,000
17,000 270,000
300 4,700
3,700 1,100
1,800 570
2 —
85
-
- -
400 _
I/ Decontamination factors used are 1.0 for Kr-85 and H-3, 1000 for the iodines
and 10^ for the actinides.
-------
11
nationally deposited tritium rather than from the worldwide distribution
of tritium. However, most of the United States population dose from
krypton-85 is due to the worldwide distribution of this isotope.
Local and regional population doses could vary considerably due to
differences in the population distributions and. .meteorological
conditions around specific plants as well as variations, in efficiency of
effluent control equipment. However, estimated national doses are
probably conservative. The projected world doses are probably correct
within a factor of five unless decontamination factors change by large
amounts.
The estimated health effects (in terms of cancer induction and/or
mortality) for a single 5 MTU/day reprocessing plant operating over 40
years are given in table 3. These values are based on the population
doses in table 2 and dose-effect conversion factors presented in
appendix C. The dose-effect conversion factors are based on a recent
study performed by the National Academy of Sciences and use the
assumption of a linear non-threshold dose-effect relationship. This
assumption is considered prudent for decision making. The health
effects were projected on the basis of the total dose irreversibly
committed by environmental releases, and include the effects of
extremely long-lived radionuclides for the first 100 years. The
individual risk calculation assumes that an individual resides 40 years
at a location 3 kilmeters from a reprocessing plant and also obtains his
food and water from this location.
-------
12
TABLE 3
Projections of Total Health Impact
Estimated Induced Health Effects for 40 Years of Operation
Radionuclide Critical
Organ
Kr-85 Whole-body
Lung
Skin
Gonad^
H-3 Whole-body
Goaads^/
1-129 Thyroid-
Individual
(? 3 km
6.0xlO-6
l.SxlO-6
1.5X10""6
6.0xlO~6
5.2xlO~5
5.2xlO~5
Regional
Population
0.38
0.095
0.095
0.17
3.3
1.2
.012
United States
Population
8.3
2.1
2.0
3.6
59
22
.024
World
Population
130
32
32
57
18
6.9
, Total
140^
34^
*y
^
sol/
30 £ /
.04^
infant
7.4x10
Thyroid- ,Q2
adult
1-131 Thyroid- .12
lQfant 4.5x10 -6
• Thyroid- .04
adult
9 /
Actinides Lung 2.0xlO"6 0.13 0.4 - 0.5
1. 50% Mortality
2. Very high Mortality
3.. Low or Zero Mortality
4. Probably less than 25% Mortality
5. Genetic affects only.
-------
13
About two-thirds of the health effects are estimated to accrue to the
world population outside the United States, but for this group, the indi-
vidual dose rates are extremely small (much less that one millirem per year).
The regional population group incurs about 25% of the total effects, but
receives larger individual doses than experienced by the world population group,
CmiULATIVE RADIOLOGICAL IMPACT .OF THE INDUSTRY
In addition to determining the impact of individual plants, the
radiological impact of the entire United States nuclear fuel reprocessing
industry was assessed. This total industry impact assessment considered
the total accumulation of the long-lived isotopes and the health effects
which can be attributed to them, A determination was made of the total
environmental buildup of long-lived environmental radionuclides resulting
from operations of the United States industry through the year 2020.
These accumulations are shown in figure 1 and represent the estimated
cumulative environmental inventories of tritium, krypton-85, iodine-129,
and plutonium-239. These inventories determine the magnitudes of doses
and of future health effects resulting from such cumulative environmental
contamination. ,
Health effects attributable to the presence of these radioisotopes in
the environment were considered in terms of the cumulative health effects
that will be caused in the future due to release of these isotopes during
one year of operation of the entire industry, as well as for the
accumulated inventories shown in figure 1.
Figures 2 through 5 represent the estimated health effects that will
be committed by the environmental buildup of certain long-lived
radionuclides if the industry is allowed to operate through a given year
at the plant decontamination factors typical of current design (1.0 for
-------
10
14
10
Figure 1. Cumulative environmental buildup of
radioisotopes released from total
fuel reprocessingindustry
10'
10
10
CURIES
10
10'
10
10
I
Df=1
10
10
10
CURIES
co
i 1.01
1970 1980 1990 2000 2010
YEAR
2020
-------
15
400
Figure 2, Estimated past and future health effects
committed by tritium releases from the
United States fuel reprocessing industry
OS.
00
(_
CJ
1980
1990
2000
2010
2020
YEAR
-------
16
Figure 3. Estimated past and future health effects
committed by krypton-85 releases from the
United States fuel reprocessing industry
7000 -
1980
1890 2000
YEAR
2010
2020
-------
17
3.0
Figure 4. Estimated past and future health effects
committed by iodine-129 releases from the
United States fuel reprocessing industry
FUTURE LIMITED TO 100 YEARS
OS
CNI
1 2.0
€/>
1,0
1S70 1980 1990 2000
YEAR
2010
2020
-------
18
Figure 5. Estimated past and future health effects
committed by aetinide releases from the
United States fuel reprocessing industry
30 -
FUTURE LIMITED TO 100 YEARS
o
3C
I—
CJ
««
C9
I— .
Z&
ea
t^>
(_
CJ
20
10
1970
2020
-------
19
tritium and krypton-85, 103 for todine-129 and 10 for the actlnides).
For example, in figure 2 for tritium, if all tritium is released from
fuel reprocessing plants with a decontamination factor of 1.0, then by
the year 1990, the graph Shows that three estimated health effects will
have been committed by exposures received prior to 1990 and nine
estimated health effects will be caused by exposures beyond 1990 from
the quantity of tritium already in the environment in 1990. By the year
2020, 105 health effects will have been committed by past exposures to
tritium and 250 estimated health effects will be caused by future
exposures to past releases.
Future health effects for tritium and krypton-85 are estimated
assuming the complete decay of the quantities present. For iodine-129
and the actinldes (238Pu, 239Pu, 2lf°Pu, 241Pu, 2tflAm, 2MfCm) only the
first hundred years of exposure beyond the time of interest are considered
for estimating health effects. The calculational techniques used are
described in appendix Di
These curves demonstrate a rapid change in the environmental impact
of the fuel reprocessing industry, especially after the 1980's. They
indicate that if reprocessing plants continue to release radioactive
material, especially krypton-85 and tritium, at current levels of
emission (which are well below current regulations), an environmental
burden of radioactive material will accrue which presumably could result
in a significant number of avoidable health effects. Iodine-129 and the
actinides have extremely long half-lives, and could impose additional
-------
20
health risks for future generations if they remain in the biosphere
beyond the 100 years assumed for this analysis,
ECONOMICS OF HEALTE EFFECTS REDUCTION AHD EFFLUENT CONTROL
The following economic considerations apply to a single
representative 5 tonne/day fuel reprocessing plant. Industry totals, as
applicable, are obtained by multiplying by the number of projected
plants (appendix A).
The fractional cost of fuel reprocessing in terms of total energy
cost was derived. At present, the value of plutonium is approximately
$7-S/gram while the value of uranium, in the form of oxide, is about
$.Q2/gram. For a typical 1,000 MW(e) nuclear power plant, costs of all
aspects of the fuel cycle from mining through waste disposal represent
about 18% of the total costs of power (Ref. D-16). The reprocessing
portion alone represents about 7% of the total costs of the fuel, or
1.3% of the cost of power generation (Ref. D-16). The capital
investment for the Allied-Gulf Company's Barnwell Fuel Recovery Plant (5
tonne/day capacity for U02 and U02 + PuOa fuels) is approximately $80
million. A plant of this size is capable of processing fuel from
approximately 40 to 50 power reactors, each of which cost $300 to 400
million. Since the only reason for fuel reprocessing is to support
electric power generation by reactors, it is apparent that neither
reprocessing nor substantial emission control cost could significantly
alter the cost of power production.
Operating costs for a 5 tonne/day plant have been estimated to be
about $13 million per year when operating at full capacity. Of this,
-------
21
approximately $3 million is for labor. State and local taxes, insurance
and interim replacement can give rise to an estimated additional annual
expense of about $2.4 million. The capital costs associated with
environmental studies, research, and construction is reported to
approach 301 of the capital costs of the plant, much of which has
already been incorporated in plants appropriate for reprocessing fuels
used in the uranium cycle. For the plutonium cycle, the level of
investment for environmental considerations is anticipated to be about
twice that for processing of uranium-cycle fuel. Present effluent
treatment capital costs are estimated to be approximately 6% of the
total capital costs.
fable 4 summarizes, for a model plant, the estimated health effects
to the world's population attributable to 40 years of operation and the
total costs of waste control systems for the four isotopes considered in
detail by this analysis. These total costs were computed assuming an
effective interest rate of 24%, a debt lifetime of 20 years and the
operating costs of the system over the projected life of 40 years. A
7.5% discount factor was used to estimate the present worth of these
costs. From these values, the cost effectiveness of the individual
component systems for control of emissions have been calculated. The
health effects reduction as a function of system cost is displayed in
figure 6. From these data, the overall cost effectiveness of combined
systems is generated and is displayed in figure 7 for health effects to
the. worldls population.
-------
Radionuelide
TABLE 4
Summary of Health Effects and Costs of Emission. Controls
for a 5 tonne/yr Fuel Reprocessing Plant
Removal System
Total Cost
(present worth)
FILTER
a ^ 60% mortality
b * 63% mortality
c Probably less than 25% mortality
d *> 100% mortality
Health Effects
Averted
Effects/$106
Krypton
Tritium
Iodine
Actinides
Cryogenic
Distillation
Cryogenic
Adsorption
Freon Msorption
Voloxidation (est)
Caustic Scrubber
Silver Zeolite
HEPA Filters
Sand Filter and
$ 11.5 million 270a
12.1 million
6.4 million
31 million 110b
3.5 million 303°
1.85 million 34
1 million 530d
2.1 million
23.5
22.3
42.2
3.5
86.6
18.7
530
260
NJ
-------
23
S
NUCLEAR FUEL REPROCESSING - 40 YEAR OPERATION
5 TONNES/DAY THROUGHPUT
UJ
1200-
1000-
80O-
600-
4OO-
200-
World Population Group
at Risk
Total Effects Avoided
Projected Mortalities Avoided
COMPONENT CONTROL SYSTEMS
ADDED IN ORDER OF
DECREASING COST EFFECTIVENESS
COMPONENT CONTROL SYST.EMS
P - PART1CULATE T - TRITIUM
I - IODINE R - RESIDUAL
K - KRYPTON
O 10 2O 3O 40 50
CONTROL SYSTEM COST
(106 Dollars)
FIGURE 6. RISK REDUCTION vs CONTROL SYSTEM COST
60
7O
-------
24
550-
5OO-
450-
tn
UJ —
^ Q 400-1
U^
uj ;=
P
i
NUCLEAR FUEL REPROCESSING - 40 YEAR OPERATION
5 TONNES/DAY THROUGHPUT
COMPONENT CONTROL SYSTEMS ADDED
P - PART1CULATE T - TRITIUM
I- IODINE R - RESIDUAL
K - KRYPTON
200-
LU
>-
1/5
100-
50-
0
....
*
COMPONENT CONTROL SYSTEMS
ADDED IN ORDER OF INDIVIDUAL
DECREASING COST EFFECTIVENESS
r^n
x\t
T
World Population Group
at Risk
Total Projected Effects Avoided
Projected Mortality
Avoided
._|—/?-
0
10
60
20 30 4O 50
CONTROL SYSTEM COST
(106 Dollars)
FIGURE 7 - CONTROL SYSTEM COST EFFECTIVENESS vs CONTROL SYSTEM COST
7O
-------
25
Certain waste product removal systems are already incorporated in
present plant designs. These include provisions for filtration of
particulates, including the actinides, and for reduction of iodine
emissions. As is shown by the figures, krypton removal systems are
almost equally as cost effective as control systems for iodine which are
currently incorporated into plant designs when viewed, in terms of the
worldwide population, Tritium removal systems are less cost effective,
on the same basis.
Some of these conclusions may be modified as the underlying bases
are changed. For example, if the calculations are based on mortality
rather than total health effects, the relative cost effectiveness of
iodine and krypton removal systems are interchanged. Similarly, if the
calculation is based on regional effects rather than world effects, the
cost effectiveness of all systems except that for control of actinides
is sharply reduced. These results are described in more detail in
appendix D.
SUMMARY AMD CONCLUSIONS
The foregoing analysis of the potential environmental impact of the
fuel reprocessing industry, and of the feasibility of minimizing this,
has involved consideration of:
1. projection of nuclear power demand and associated spent fuel
inventories expected to be reprocessed.
-------
26
2. the present and potentially available technology for fission
product removal including estimates of costs based on initial cost
and operating expense,
3. the distribution throughout the environment of certain
radionuclides released during normal operation and the resultant
doses to the regional, national, and world populations, and
4. estimates of the statistical relationships of population
exposure for specific radionuclides and the number of health effects
expected to be associated with these.
The major conclusions derived from these considerations are as follows;
1. the fuel reprocessing industry, as an intergral part of the
entire nuclear power industry, can be treated as a separate entity
for purposes of evaluating its contribution to the overall release
of radionuclides to the environment,
2. the quantites of certain radionuclides which may be released, if
industry growth projections are substantially correct, are large.
Consequently, the reprocessing step represents the point in the
nuclear fuel cycle where the consequences of release of the long- •
lived isotopes, such as tritium, krypton-85, iodine-129, and
Plutonium 239 should be carefully considered,
C;
-------
27
3. the consequences of the buildup of very long-lived radionuclides
(such as iodine-129 and plutonium-239) from the fuel reprocessing
industry can produce cumulative and irreversible environmental
levels which can be projected to cause adverse health effects on a
national and worldwide scale. The consequences of the release of
the shorter-lived isotopes, while possibly more severe on a short-
term scale, are more easily reversed by effective control
techniques,
4. control technology exists to reduce emissions of these materials.
During the near-term future, all the tritium and krypton-85 produced
during the fission process will be released to the environment
during the reprocessing stage. Some control currently is applied
for iodine and plutonium releases. Control technology to achieve
confinement of krypton is essentially developed and some reduction
for nearly all radioisotopes would appear feasible on a long-term
basis,
5. removal of plutonium and other actinides is the most cost
effective method of health risk reduction. Next in order of
effectiveness are systems for removal of krypton and iodine which
are about equal from a cost effectiveness standpoint. Tritium
control technology is least cost effective at the present time, but
future developments should be pursued to alleviate this problem,
-------
28
6, available data allows an estimate of the incremental additional
cost to the cost of nuclear production of electricity from the
imposition of waste control systems at fuel reprocessing facilities
to be about 0.1 percent of the total present cost to the producer of
electricity,
7. evaluation of the total environmental impact of radioactive
effluents requires a consideration of dose commitments beyond those
delivered immediately in the site area, and, because of the long life
of those materials in the biosphere, must include the exposures to
national and world populations, beyond those delivered in the year
of release, and
8. although the estimated health impact from the operation of a
single 5 tonne/day reprocessing facility is relatively small, ,
extrapolation of the industry as a whole over the next 50 years of
operation indicates that long-term cumulative effects may be quite
large4
-------
29
REFERENCES
1. Nuclear Fuel Services, Inc. Safety Analysis Report, AEC
Docket No. 50-201, July 1962.
2. Kelleher, W.J., Environmental Surveillance Around a Nuclear
Fuel Reprocessing Installation, 1965-1967, Radial Health
Data Rep., 10, August 1969,
3. Logsdon, J.E., and J.W.N. Hickey, Radioactive Waste
Discharges to the Environment from a Nuclear Fuel
Reprocessing Plant, Radiol Health Data Rep., 12, 6,
July 1971.
4. Shleien, B., An Estimate of Radiation Doses Received by
Individuals Living in the Vicinity of a Nuclear Fuel
Reprocessing Plant in 1968, BRH/NERHL 70-1, May 1970.
5. Magno, P., et. al., Liquid Waste Effluents from a Nuclear
Fuel Reprocessing Plant, BRH/NERHL 70-2, November 1970.
6. Cochran, J.A., et. al., An Investigation of Airborne
Radioactive Effluent from an Operating Nuclear Fuel
Reprocessing Plant, BRH/NERHL 70-3, July 1970.
7. Smith, D.G., et. al., Calibration and Initial Field Testing
of 85 Kr Detectors for Environmental Monitoring, BRH/NERHL
70-4, November 1970.
8. General Electric Company, Midwest Fuel Recovery Plant Safety
Analysis Report, AEC Docket No, 50-332, September 1969.
9. Allied Gulf Nuclear Services, Barnwell Nuclear Fuel Plant
Safety Analysis Report, AEC Docket No. 50-332, September 1969
10. Rogers, W.M. - Nuclear Fuel Reprocessing Requirements in the
Western United States, Western Interstate Nuclear Board,
December 1971«
11, Oak Ridge National Laboratory, Siting of Fuel Reprocessing
Plants and Waste Management Facilities, ORNL-4451, July 1970
-------
-------
APPENDIX A
SPENT NUCLEAR FUEL RADIOACTIVITY FORECASTS
-------
-------
A-l
I. INTRODUCTION
As a starting point In the assessment of the radiological
impact of nuclear fuel reprocessing plants on the general
population, an estimate of the total quantity of radioactive
materials present in spent fuels produced by nuclear electric
power generation must be obtained.
These estimates are based primarily on the projected
electric power demand and on the fraction of that demand
expected to be satisfied by nuclear plants. To some extent,
they are also contingent on the projected number of various
reactor types resulting in differing amounts of the various
radionuclides produced. The annual reprocessing radioactive
waste and product inventories for specific nuclides are
forecast through the year 2020.
II. SIZE AND SCOPE OF THE U.S. INDUSTRY
Electrical power demand forecasts for the United States, and
the fractional amount expected to be produced by the nuclear
industry through the year 2020 as used in this study, are
summarized in table A.I (Ref. A-l and A-2),
The isotopic composition of nuclear fuels is expected to
change over the next several decades. Presently uranium-235
is the most widely used reactor fuel material. However, the
expected increase in energy demand, coupled with the inefficient
utilization of the fissionable material by reactors as currently
designed, would deplete the available low-cost natural uranium
resources, which contain only about 0.7% uranium-235, by the
-------
Nuclear electric
Year generation:GW(e)
TABLE A.I
Estimated U.S. Fuel Reprocessing Requirements
(Adapted from References A-l and A-2)
*
• Tonnes of fuel discharged annually
.Number of 5 tonne/day
reprocessing plants
required
1970
1975 - '
1980
1985
1990
1995
2000
2005
2010
2015
2020
2.6
. . 40
110
220
420
650
1000
1360
1780
2220
2700
LWR-U
25
700
1900
2700.
3700
4100
3700
3700,
4300
5300
6100
LWR-PU
0
90
500
2600
3800
4100
.- 3800
3700
4400
5400
,. 6100
LMFBR
0
0
1
0
480
2,600
'll.SOO
20,600
32,800
43,800
58,000
HTGR
0
0
1
100
2,420
6,600
. 7,800
10,000
10,000
10,000
8,800
TOTAL
25
790
2,400
5,400'
10,400
17,400
26,800
38,000
51,500 .
64,500
79,000
1
' 1
, ' 2
4
E
12
18
26
35
43
53
Burnup: 33 GWd(t)/tonne and 0.35 thermal efficiency.
-------
A-3
end of the century. Other fissionable materials, plutonium-
239 and -241 and uranium-233, are then increasingly expected to be
used to meet long-term power requirements since these can be
produced as byproducts of reactors generating electric power;
i.e., plutonium-239, -241 from uranium-238 and uranium-233 from
thorium-232.
The types of reactors are also expected to be altered as
technology advances. Currently, in the United States, most
nuclear power plants use light-water-cooled reactors (LWR) of
two types, pressurized water reactors (PWR) and boiling water
reactors (BWR), These are fueled with natural uranium slightly
enriched to give an isotopic composition of approximately
3% uranium-235 and 97% uranium-238. In the future, many of
these LWR systems are expected to be partially fueled
with recycled plutonium instead of uranium-235.
Light water reactors are inefficient producers of plutonium;
the ratio of the amount of fissionable plutonium produced to
the amount of fuel consumed (the conversion ratio) is about 1:3.
•More efficient reactor types, namely converter reactors which
produce nearly as much fuel as is used, and breeder reactors,
which produce more fuel than is used, are expected to become
a significant part of the nuclear power industry after the
year 1985. Examples of these two reactor types most likely
to be used are the high temperature gas-cooled converter
reactor (HTGR) fueled with uranium-233 and uranium-235, and the
liquid metal fast breeder reactor (LMFBR) fueled with.
plutonium.
-------
A-4
Knowledge of the relative numbers of each type of reactor
is important primarily for determination of the total quantities
of the actinides produced per unit of electric power generation.
Fission yields for most of the other elements of interest remain
nearly constant for all fissionable materials of interest. For
this study, a mix of reactor types as given in a study by, the
AEC (Ref. A-l) was used. Power capacity values were converted
to power generation values by using a 64% load factor
(percent of maximum available power utilized). These were
then converted to tonnes of fuel discharged in any given year
by using data on power generated two years earlier, a thermal
efficiency of 0.35, and a burnup of 33 GWd per tonne of fuel:
tonne of fuel discharged per year - (gigawatt (GW))
.^ •> tn f, power generated, ,365 days,
power capacity) x (0.64 •*• * rr—) x ( J~~) x
r power capacity year
JL _ Thermal power (t) •> ,1 tonne fuel ,
XJ.35 electrical power (e); x (33 GW days (t)''
A sutnmary of spent fuel projections by reactor type and
total amounts is given in table A.I. The number of fuel
reprocessing plants necessary to service the nuclear power
industry was estimated by using these spent fuel projections
and the assumption that each fuel reprocessing plant will
handle 5 tonne of spent fuel per day (equivalent to
1500 tonne/year). These numbers were also summarized in
table A.I for 5-year increments.
-------
A-5
The uncertainty in the above total nuclear electric
power generation values and total spent fuel discharge values
after 1980 is estimated to be less than a factor of two. The
distribution by reactor types may have a somewhat larger
uncertainty.
III. QUANTITIES OF RADIONUCLIDES IN SPENT REACTOR FUEL
There are three types of radioactive material present
in spent reactor fuel: fission products, activation products,
and actinide isotopes. The quantities of specific radionuclides
present are primarily determined by fuel type, amount of
burnup, and time of cooling (time between removal from the
reactor and reprocessing).
Tables A. 2 and A.3 show quantities of the potentially
significant fission product and activation radionuclides present
in one tonne of spent fuel with 33 GW(t)-days burnup and 150 days
cooling time. These values (adapted from Ref. A-2) are considered
reasonably representative of all nuclear fuel types. There is
indication that cooling times shorter than 150 days may be used for
some fuel cycles in the future, since quicker recycling of the
recovered fuel produces an economic benefit. This would
significantly increase the amounts of shorter-lived radionuclides
in the fuel and available for release, but would not affect the
long-lived fission product inventories.
The amounts of actinides estiniated to be present in
uranium fuels and plutonium-recycle fuels are given in
-------
TABLE A.2
Representative Quantities of Potentially Significant Fission Products In Spent Reactor Fuels
(Adapted from Reference A-2)
Isotope
H -3
Kr -85
fc -99
Ru -103
106
Te -125m
127m
129m
I -129
131
Cs -134
135
137
Sr -89
90
Y -91
Zr -93
-95
Nb -95
Sb -125
Ce -141
144
Pm -147
Eu -155
Half-life
(Years)
12.3
10.7
2. 13x10 5
0.11
1.01
0.16
0.30
0.09
17x10 6
0.02
2.05
3x10 6
30.2
0.14
28.9
0.16
0.95xl06
0.18
0.10
2.73
0.09
0.78
2.62
5.0
Curies per
tonne
800
10,500
15
180,000
820,000
6,500
25,000
13,000
0.04
2.0
100,000
1.2
106,000
100,000
60,000
190,000
2
400,000
800,000
13,000
80,000
800,000
200,000
40,000
Grams per
tonne
0.083
27
880
5.7
240
0,36
2.7
0.42
250
0.01
77
1400
1200
3.5
430
7.8
490
19
21
12
2.8
250
220
87
Release
state
Gas
Gas
Semivolatile
Semi-volatile
Semi volatile
Semi volatile-
Semi volatile
S emi vo la t i le
Volatile
Volatile
Semi volatile
Semivo lat i le
Semi volatile
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Solid
Notes
95% released as HTO
Oxide b.p. 200 °C
Tetroxide b.p. ,80°C
169Rh + l°6Rh daughters
Oxide b.p. 750 °C
Oxide b.p. 127 fe daughter
Oxide b.p.1 29 Te daughter
b.p. 184°C
b.p. 184°C
Oxide b.p. 750 °C
Oxide b.p.
Oxide b.p. 75Q°C137™Ba daughter
90Y daughter
95ffiNb + 95ub daughters
l«rtpr + IWm daughters
Burnup = 33 GWd(t)/tonne
Cooling Time = 150 days
-------
A-7
TABLE A. 3
Representative Quantities of Potentially
Significant Activation Products in Spent Reactor Fuels
(Adapted from Reference A-2)
Isotope
Mn-54
Fe-55
Fe-59
Co-58
Co-60
Half-life
(years)
0.86
2.7
0.12
0.20
5.26
Curies per
tonne
30,000
20,000
500
30,000
2.000
Grams per
tonne
3.9
8.3
.01
1.0
1.8
Release
stage
Solid
Solid
Solid
Solid
Solid
Based on 33 GWd(t) burnup/tonne
150 days cooling time
-------
A-8
table A,4 (Ref. A-2 to A-8). It was assumed that all fuels
(including those used in HTGR's) other than uranium-235 fuels,
can be considered equivalent to Plutonium-recycle fuels.
This introduces an uncertainty in these values proportional
to the deviation from this assumption.
Based on the amounts of spent fuel to be processed, and
on the estimated quantities of radionuclides per tonne of
spent fuel, the projected annual quantities of several of
the most significant radionuclides in fuel to be processed
were calculated and are presented in table A. 5,
Based upon plant decontamination factors appropriate for
control systems presently incorporated in fuel reprocessing
Q 0
facilities [10 , 10 , 1 and 1 for airborne pollutants (actinides,
iodines, krypton, and tritium, respectively)], the annual inventories
shown in table A. 5 can be used to project the buildup of long-lived
radionuclides in the environment as a result of fuel reprocessing.
The estimated buildup of tritium, krypton-85, iodine-129, and
plutonium-239 are shown in figures A.I through A.4. Estimates of the
environmental levels for time periods beyond 2020 by extrapolation
of these curves is inappropriate. Detailed discussion of the
status of system decontamination factors is presented in
appendix B.
-------
A-9
TABLE A. 4
Representative Quantities of Actinides Present in Spent Reactor Fuels
(Adapted from References A-2 to A-8)
Isotope
U-235
236
238
Np-237
Pu-238
239
240
241
242
Am-241
243
Cm-242
244
TOTAL
(excluding
Half -life
(years)
710x10 6
24xlQ6
4 5 10x10 8
2x10 6
86
24,400
6,580
13
379,000
458
7,800
0.45
17.6
uranium)
Uranium fuels
Ci/tonne
<1
<1
<1
<1
4,000
500
650
150,000
2
750
20
35,000
2,000
193,000
g/ tonne
8,000
4,000
950,000
600
230
8,100
2,900
1,300
510
230
100
10
25
14,000
Pu-recyele
Ci/tonne
<1
<1
<1
<1
6,000
750
1,000
300,000
5
2,000
200
250,000
25,000
585,000
fuel
g/ tonne
3,000
1,500
950,000
200
340
12,000
4,400
2,600
1,300
620
1,000
75
300
23,000
Cooling time = 150 days
Burnup =33 GWd(t)/tonne
-------
TABLE A,5
£
Estimated Annual Inventories of Selected Nuclides in Spent Reactor Fuels
(Curies)
Year
1970
1975
1980
1985
1990
1995
2000
2005
2010
2015
2020
Fuel Discharge
(IfC)
25
790
2,400
5,400
10,400
17,400
26,800
38,000
51,500
64,500
79,000
Tritium
2.0x10*
6.3xl05
1.9xl06
4. 3x10 6
8. 3x10 6
1.4xl07
2. 1x10 7
3. 0x10 7
41. 1x10 7
5. 2x10 7
6. 3x10 7
Krypton-85
2. 6x10 5
8. 3x10 6
2. 5x10 7
5. 7x10 7
1.1x10 8
l.SxlO8
2. 8x10 8
4. 0x10 8
5.4x10 8
6.8x10 8
8. 3x10 8
Iodine-129
1.0
3.2x10 l
9. 6x10 l
2.2X102-
4.2xl02
J.OxlO2
l.lxlO3
l.SxlO3
2. 1x10 3
2. 6x10 3
3.2xl03
Plutonium-239
1.9x10*
5. 9x10 5
l.SxlO6 -
4.2xl06
7.8xl06
1.3xl07
2. 0x10 7
2. 9x10 7
3. 9x10 7
4.8xl07
5.9xl07
Plutonium-241
7.5xl06
2. 4x10 8
' 7. 2x10 8
1.6xl09
3.1xl09
5.2xl09
i
S.OxlO9 o
l.lxlO10
l.SxlO10
1.9xl010
2.3xl01Q
Based on plutonium-recycle fuel and reactor type distribution in table A.I
(33 GWd(t)/tonne burnup and 150 days cooling period.)
-------
A-ll
700
600
500
400
D
U
O
300
200
100
0
1970
1980
1990 2000
Year
2010
2020
Figure A.I Estimated Cumulative Environmental Build-up of Tritium
from the Fuel Reprocessing Industry in the United States using
a Decontamination Factor of 1.
-------
A-12
8000
7000
6000
5000
3
O 4000
D)
-------
A-13
0
1970 1980 1990 2000 2010 2020
Year
Figure A.3 Estimated Cumulative Environmental Build-up of lodine-129
from the Fuel Reprocessing Industry in the United States using a
Decontamination Factor of 1000.
-------
A-14
1.2
1.0
0.8
•: 0.6
3
u
0.4
0.2
0
1970
1980
2010
2020
1990 2000
Year
Figure A.4 Estimated Cumlulative Environmental Build-up of Plutonium-239
from the Fuel Reprocessing Industry in the United States using a
Decontamination Faptor of 109.
-------
A-15
A-l. Hofmann, P. L., U.S. Civilian Nuclear Power Cost-Benefit Analysis.
Fourth United National International Conference on the Peaceful Uses
of Atomic Energy, Geneva, Switzerland, 6-16 September 1971, A/CONF.
49/P/072.
A-2. OBNL Siting of Fuel Reprocessing Plants and Waste Management
Facilities. National Technical Information Service. ORNL-4451
UC-70. (July, 1970).
A-3. Deonigi, D. E. Formation of Transuranium Isotopes in Power
Reactors National Technical Information Service. BNWL-140 Rev.-l
(Jan. 1966).
A-4. Deonigi, D. E., McKee, R. W. and Haffner. Isotope Production
and Availability from Power Reactors. National Technical
Information Service. BNWL-716, Uc-23 (July, 1968).
A-5. Burch, W. D., Bigelow, J, E. and King, L. J. Transuranium Processing
Plant Semiannual Report of Production, Status and Plans for Period
Ending June 30, 1971. National Technical Information Service.
OML-4718, UC-80. pp. 29-30. (Dec. 1971).
A-6. Drumheller, K. Pacific Northwest Laboratory Division of Isotope
Development Programs Quarterly Report Nov. 1968 to Jan. 1969. National
Technical Information Service, BNWL-1010 (Feb. 1969).
A-7. Nodvil, R. J. Supplementary Report on Evaluation of Mass Spectrometric
and Radiochemical Analyses of Yankee Core I Fuel, Including Isotopes
of Elements Thorium through Curium. National Technical Information
Service WCAP-6086'. (Aug. 1969) .
A-8. Crandall, J. L. Tons of Curium and Pounds of Californium
Presented at American Nuclear Society International Meeting,
Washington, D.C. (Nov. 10-15, 1968).
-------
-------
APPEHBIX B
FUEL REPROCESSING FACILITIES
-------
-------
B-l
I. INTRODUCTION
This appendix provides a basic description of the processes used
at a fuel reprocessing plant and the control systems for reducing the
radioactive discharges. There are three commercial nuclear fuel
reprocessing plants capable of operation or under construction in the
United States. Table B.I summarizes general information for these
three facilities. Currently, these plants are not processing
irradiated fuel. The first commerical plant was Nuclear Fuel
Services, Inc. (NFS) which is shut down for expansion. The Midwest
plant is processing unirradiated fuel, and operation with spent fuel
is anticipated for the summer of 1973. The Barnwell plant is under
construction, and operations are expected to begin in 1974 or 1975.
II. GENERAL DESCRIPTION OF PROCESS
Fuel reprocessing plants are essentially complex chemical plants,
the complexity being compounded by the fact that the materials being
processed are highly radioactive. The specific process used to
separate the spent fuel element into the product streams and waste
stream is dependent upon the particular type of reactor fuel being
serviced.
All three present facilities use a shear (chop) and nitric acid
leach method to separate the light water reactor (LWR) spent fuel
from the metal cladding. Following this step, the Purex process wich
tributyl phosphate (TBP) as the solvent is used to extract the
uranium and plutonium from the fission product waste in column
contactors (Ref. B-l, B-2, B-3, and B-4). The uranium and plutonium
are separated and further purified by various means, including ion
-------
TABLE B.I
General Information for Commercial U.S. Nuclear Fuel
Reprocessing Plants
Plant
Nuclear Fuel
Services, Inc.
Location
West Valley,
New York
Owner
Getty Oil Co.
Capacity
(MTUa/year)
300
expanding to
600 to 900
Type of Process
Purex
Midwest Fuel
Recovery Plant
Barnwell Nuclear
Fuel Plant
Morris,
Illinois
Barnwell,
South Carolina
General 300
Electric Co.
Allied-Gulf 1500
Nuclear Services,
Inc.
Purex
Aquafluor process
Purex
w
NJ
MTU = metric tons uranium
-------
B-3
exchange, scrubbing, evaporation, etc. The design recovery rate is
99.5% for uranium and plutonium.
Plants designed for processing of spent fuel elements from LWR
systems could be used to process elements from LMFBR as long as the
facilities are derated (handle smaller quantities of fuel) to avoid
eritieality problems and not exceed constraints on effluents (Ref. B-
14).
The processing of HTGR spent elements requires a different "head
end" processing system since such fuel elements may require the
burning of the graphite which contains the coated fuel particles.
The decontamination of the off-gas stream resulting from the burning
operation requires development of "head end" processes unique to the
HTGR processing facility (Ref. B-15). The Thorex process (Ref. B-3)
will be used for the separation of uranium and thorium from the
fission product wastes in spent HTGR fuels (Ref. B-15).
III. PROCESS FLOW
A simplified summary of the aqueous processing steps in plants
under construction or in operation is displayed in figure B.I,
The main process steps will be summarized here to indicate sources
and handling of radioactive waste liquids and gases.
1. Cask. Unloading and Decontamination — Cask wash water is
discharged to low-level waste system.
2. Fuel Pool Storage — Special provisions are made for leaking
fuel elements storage and resulting contaminated water and off gases.
Prevention of criticality is a major design factor. Inventory
control and accountability are important operating parameters.
-------
Figure B.I Typical process flow schematic
Midwest Nuclear Fuel Reprocessing Plant
SOI
BBC
L
SHEAR .
V
TFAOff 1
1
i
f rti . . SOLVENT EXTRACTION
| 1 ACID /
1 7 Y' RECYCLE »l
PRODUCT1 RWOVTBRY >r«r«nn-t ^yy»-iyrn«Ti riT-i n rn-r/v»T
j.iiu.uuuA Miwjvjuni . WASTE CONCENTRATION
(SOLVENT RECOVERY) ' (ACID RECOVERY)
4-
tynnnrTnm _J T
101 EXCMIGE . HIGH LEVEL
A'
PRODUCT ' j —
^
II
i
I
I
— »-, ,-
1
i
PAHPICULATE
FILTERS
f 4^
Ag Z FILTER 1
„„ 1
1 ^
1 1 1
OFF GAS
SCRUBBER 1
1
i
I
tJ
1
pT,nT1TTrql . URANIUM SEVERAL
TnOTTr>T< FLUORIHATIOH INPUTS i
- y
v
141 Y
UF^-
Lotdcut, ta OT ^V11
1 *" WASTE
STORAGE
^
LOW- LEVEL
^ WARW
CLADDING COHC1ITRATION
STORAGE . 1
• 'Lt
. Con
"J3~"" 0:
s
3W Level
:entrator
ff-Gas
irubber
1
i
i
i
i
i
i
j
PROCESS STREAMS
WASTE GAS STREAMS
-------
B-5
3, FuelTransfer and Mechanical, Processing ~ NFS and Midwest
disassemble fuel elements before shearing the individual fuel rods.
Barnwell will shear the entire element and thus is essentially
limited to reprocessing fuel from light water power reactors because
of the geometries involved. Lengths of sheared fuel rods range from
about 1.25 to 7.5 cm (1/2 to 3 inches). Very little of the krypton-
85 and tritium is released during the shearing process. Goode (Ref.
B-5) reported that less than 1% of the krypton-85 is released during
this step and Cochran et al. (Ref. B-6) confirmed this conclusion
based on field studies at an operating plant.
4. Fuel Dissolution — In this step the spent fuel is leached
from the sheared cladding in nitric acid as a preparatory step to the
chemical separation processes. The leached hulls are analyzed for
plutonium content and returned for further leaching if necessary.
Over 99% of the krypton-85 (Ref. B-5, B-6) and about 6% (Ref. B-7) of
the tritium is released into the off-gas system by this operation.
In addition to the noble gases, a large fraction of the halogens are
also released to the off gas in this step. Most of the radionuclide
particulates present in the off gas result from this step (Ref. B-8),
although this may not be the case at Midwest where large quantities
of particulates will be added from the high-level waste
solidification process. Since NFS dissolves fuel on a batch basis,
essentially all the krypton-85 in the batch is discharged within a 3-
to 4-hour period. Present NFS Technical Specifications limit fuel
dissolution to 2 metric tons per day. Midwest and Barnwell will
dissolve on a semieontinuous basis. Waste gases are processed
-------
B-6
separately through the dissolver off gas system (DOG) at NFS and
Barnwell, since most of the radionuelides in the waste gases are
generated in this step at these facilities. Midwest has designed
their gaseous waste treatment system somewhat differently and waste
gases are not segregated by source.
- 5, Chemical Separation and Purification — The nitric acid feed
from the dissolver which contains the uranium, plutonium, and fission
products is counterflowed in a contactor column with TBP and nitric
acid. The uranium and plutonium are preferentially dissolved in the
TBP, and the fission products are retained in the nitric acid.
Separation of the plutonium and uranium in the organic solvent is
achieved by reducing the plutonium to its trivalent state where it
can be stripped with a nitric acid scrubbing process in a contactor
column. The uranium is subsequently also stripped from the TBP which
is then recycled. The plutonium and uranium are then purified
through a series of processes. Midwest has designed a calcining
system for conversion of the uranium to UF for shipment directly to
an enrichment facility. NFS and Barnwell will ship uranyl nitrate in
tank trucks to conversion facilities. Plutonium will be stored and
shipped in the nitrate form in critically safe containers.
6. Recovery of Solvent and Acid — The TBP solvent and the acid
are recovered and recycled. The fission products are concentrated in
the evaporator bottoms of the high-level waste concentrator system
and the acid overheads are recovered. Additional fractions of the
semivolatiles and the halogens are discharged to the waste gas system
in this step, along with some particulates.
-------
B-7
7. High Level Waste -- At MFS and Barnwell high-level liquid
waste will be stored in stainless steel tanks which are located
underground and externally cooled. These tanks are placed on
concrete saucers to permit monitoring for leaks and inside concrete
vaults which provide a secondary containment. These acidic wastes
are not neutralized prior to storage to simplify future
solidification. At the Midwest facility, which uses a fluidized bed
calcination process for solidification, high-level wastes are
solidified immediately after separation. This process has been used
on a production scale with intermediate level waste since 1963 at the
Idaho Chemical Processing Plant (ICPP) (lef. B-9). The solidified
waste will be sealed and stored in a cooling pool. The off gases
from this process will contain some radioactive particulates (Ref. B-
8, B-9), especially the semivolatile fission products such as
ruthenium.
8. Low Level Waste — The low-level liquid waste stream consists
of wastes collected from sources throughout the plant. Midwest and
Barnwell have been designed to completely eliminate the discharge of
low-level liquid waste by adding an evaporator to the system to
process the final low-level liquid waste stream (it is discharged to
the environment at IIS). The evaporator overheads are discharged
through the stack and theoretically contain all remaining tritium
from the fuel. The bottoms are solidified and are currently shipped
to privately operated waste burial areas on site.
In the gaseous stream, almost all of the krypton-85, and varying
fractions of tritium and other volatile nuclides, such as iodine, are
-------
B-8
released to the dissolver off gas (DOG) during the dissolution
process. Treatment of this waste stream is complicated since it
contains varying concentrations of the oxides of nitrogen. NFS and
Barnwell are designed to have a separate waste system for the DOG.
The various process tanks have vents for off gases which are
contaminated with volatile fission products. These are routed to the
vessel off-gas system (VOG). Midwest will combine the DOG and VOG
systems since they have additional sources of airborne wastes from
the calcining of the high level liquid waste.
The gaseous waste treatment system at all three facilities
basically consists of a caustic scrubber, followed by a silver
zeolite absorber and then final filtration through a high efficiency
particulate air (HEPA) filter. NFS has incorporated both an acid
scrubber and a caustic scrubber in their system. The scrubbers and
silver zeolite adsorption systems are installed to collect the iodine
in the off-gas stream. Midwest and Barnwell have installed or plan
to install in series two independent particulate filtration systems
which are isolated to avoid dual failure. Two HEPA filter syterns are
to be installed at Barnwell. Midwest has installed a HEPA filter
system in the off-gas stream and a sand bed filter system for final
filtration of both the off-gas stream and the plant ventilation air.
IV. CONTROL POINTS FOR EFFLUENTS
The cladding on the fuel normally provides the primary barrier
(or containment) for preventing the release of fission product
wastes. This barrier must necessarily be destroyed in the nuclear
fuel reprocessing plant in order to recover the fissile and fertile
-------
B-9
material for reuse. This fact, in addition to the very large
quantities of fission products present, requires that effluent
control procedures must be incorporated at all processing steps. In
practice, control of the effluents from each process can be achieved
through the use of a common collecting system such as the vessel off-
gas header system which collects the off gases from several
processing vessels. The use of such common systems has definite
economic advantage. Fewer control systems are required and a higher
degree of reliability for control of discharges is possible since
there are fewer components subject to failure, better quality
equipment can be installed.
An example of a common collection system is the low-level liquid
radioactive waste system. Sources feeding this system include, but
are not limited to: cask decontamination water, leakage from the fuel
storage pool and waste storage tanks, laboratory wastes, laundry
wastes, high-level waste equipment drains, and floor drains. This
waste is collected and processed through an evaporator where the
overheads can be condensed and recycled, discharged, stored, or
handled in a combination of these options. NFS, Inc. has chosen the
method of discharge of the condensed overheads through a system of
settling lagoons to a public waterway. Midwest has chosen a
combination of treatments where the evaporator overheads are
condensed and recycled to the maximum extent possible. Low-level
liquid wastes which cannot be recycled, such as air scrubbing wastes,
are collected in a low-level waste vault which is maintained at a
constant volume by use of a second evaporator. The overheads of this
-------
B-10
evaporator are not condensed but discharged through a second air
cleaning system and the sand filter to the stack.
In general, air cleaning systems follow the same principle.
Contaminated air is collected from various processes in a common
header and then treated to remove contaminants. Reprocessing plants
have sufficient chemical contaminants in some off-gas streams to
cause problems, such as overloading of various air cleaning systems.
Therefore, the off-gas streams are frequently segregated by source,
especially for initial treatments.
Specific effluent control points and the principal contaminants
are; (1) Dissolver Off Gas — noble gases (krypton-85 and xenon-133),
halogens (iodine-129 and iodine-131), tritium, and particulates. (2)
Solvent Extraction and Purification Off Gas — particulates. (3)
Solvent Recovery Off Gas — mixed fission products. (4) Acid
Recovery Off Gas (High-Level Waste Concentration) — halogens and
other volatile species. (5) High-Level Waste Solidification —
ruthenium and other potential volatile radionuclides including
technetium, cesium, selenium, and tellurium (Ref. B-8). (6) Low-
Level Waste Treatment — tritium, strontium, ruthenium, cesium, and
other longer-lived radionuclides (Ref. B-10).
Investigations are being conducted into modifying spent fuel
reprocessing systems to provide more positive control of the
effluents. ORNL is presently directing efforts to the design of
equipment and process flow to obtain "near zero release" processing
of short-cooled LMFBR fuel (Ref. B-ll). The design includes a new
"head-end" processing step (voloxidation) that is designed to release
-------
SECONDARY
GAS
TREATMENT
MJT3
FINAL
OFF-GAS
TREATMENT
mjr>TO *TTJ
TO STACK
SPENT
FUEL
RECEIVING
THROUGH
VQLOXIDATION
DISSOLUTION
THROUGH
FEED
ADJUSTMENT .
SOLVENT EXTRACTION
THROUGH PRODUCT
CALCINATION
HN03- H20
RECOVERY
PuO,
HNO
RECYCLE
3*"1
H2°
WASTE
SOLIDIFICATION
AND SOLID WASTE
PACKAGING
STORAGE
Figure B.2 Schematic diagram showing the steps required for the
"zero release" reprocessing concept (B.ll)
-------
B-12
tritium from WFBR fuel and deactivate sodium prior to aqueous
processing. The tritium, krypton, and iodine are evolved during the
head end operations (which include voloxidation and dissolution) and
vented to their respective primary removal systems. The off gas from
the dissolution, cell and from the process equipment beyond
dissolution and feed adjustment is subjected to a secondary off-gas
treatment system which includes filters for partieulates and
scrubbers for the oxides of nitrogen, halogens, and ruthenium,
Argonne National Laboratory is also developing an alternate "head-
end" pyrochemical process for decladding of LMFBR fuel which
potentially may produce an improved method for control of the
effluents (Ref. B-12, B-13).
V. SOLID WASTES
The solid wastes resulting from the recovery of uranium and
, plutonium can be categorized as high-level solidified wastes, spent
fuel cladding wastes, and low and intermediate level wastes. Federal
regulations (Ref. B-16) require that high-level wastes, generally
interpreted as self-heating, be solidified within five years of
*
processing. In addition to the radiation exposure protection which
must be provided for these wastes, cooling must also be provided.
The spent fuel cladding wastes contain residual amounts of the fuel
in addition to the activated metallic radionuelides making up the
cladding itself. There are many sources of low and intermediate
level wastes: air filters, spent resins, silver zeolite, evaporator
bottoms, sand filters, etc. It appears the greatest problem
presented by these wastes is the presence of long-lived radionuelides
-------
B-13
of health significance such as the alpha-emitting transuranics and
lodine-129. These long-lived components will be present in all solid
wastes from reprocessing plants,
VI. DISCHARGE CONTROL OPTIOMS
In general, the fuel reprocessing industry has incorporated the
most advanced technology into their waste treatment systems. For
example, the control system for iodine, one of the limiting
radiomiclides in the local environs of a reprocessing plant, (Eef, B-
17) will use silver zeolite technology which has only recently been
developed. It should be noted that a control method is not available
for tritium and only the one control system is planned for iodine.
Most of the radioactive discharges to the environment from
reprocessing plants will be in the gaseous waste Affluent. The two
newer plants have designed their processing systems to eliminate the
discharge of liquid radioactive material. However, neither of these
facilities has been operated with irradiated fuel. Thus, a decision
on whether it is preferable to discharge radioactive waste to the
atmosphere only or to discharge via both the liquid and gaseous
pathways should be postponed until operating experience with both
methods is obtained. Since most of the current efforts to reduce
discharges have been directed toward gaseous effluents, the methods
discussed in this section are limited to the control of airborne
discharges. However, liquid waste discharges may require additional
investigation in the future.
The radioactive pollutants that are most likely to be released
from normally operating reprocessing plants are krypton, tritium,
-------
B-14
iodine, and the actinides. Other fission products such as strontium,
ruthenium, and cesium, and induced activities in the fuel element
cladding can also be released. In addition to the anticipated normal
discharges (gaseous waste stream), miscellaneous airborne releases
can occur because of the complexity of the various processing oper-
ations and the unproven reliability of some of the control systems.
Such releases may not be detected by monitoring of the gaseous waste
stream (stack effluent). Inplant air monitoring for contamination
control can, however, indicate these possible pollutants. A summary
of the characteristics of gaseous waste control systems for the
isotopes of major concern is presented in table B.2. It is seen that
the long-term operational reliability of control systems is unproven
with the exception of the H1PA (high efficiency particulate air) and
sand filter systems.
. KRYPTON-85 CONTROL
Up to the present, krypton-85 and other noble gases have been
released directly to the atmosphere from nuclear reactors and fuel
reprocessing plants. With no off-gas treatment for noble gases about
10,000 Ci of krypton-85 is estimated to be discharged per metric ton
of spent fuel processed, assuming a burnup of 33,000 megawatt days
per metric ton.
Several methods have been suggested to limit such releases. The
processes are classified as ambient temperature adsorption, cryogenic
adsorption, cryogenic distillation, selective absorption,
permaselective membranes, and clathrate precipitation.
-------
TABLE B.2
Control System Data for Nuclear Fuel Reprocessing
(LWR + Recycle Fuels)
System
Decon
Projected
COSTS
Isotope
Krypton-85
Tritium
Iodine-131
Iodine-129
Aetinides
Control System
(a) None
(b) Cryogenic
distillation
(c) Cryogenic adsorp-
tion (charcoal)
(d) Freon adsorption
(a) None available
(b) Voloxidation
(a) None
(t>) Scrubber
+ AgZ
(a) None
0>) Scrubber
+ AgZ
(a) None
(b) Pre~filter
+ 2 HEPA's
(c) Pre-filter + HEPA
+ Sand Filter
Reliability
D
NA
Good
Unproven
Unproven
NA
Unproven
NA
Unproven
NA
Unproven
NA
Good
Excellent
Factor
1
a
io3
2
102
10
1 2
10
1
IO3
1
10 3
io5-io6
io9
io9
Release Rate
10 Ci/MTU
10 Ci/MTU
2
10 Ci/MTU
10 ' Ci/MTU
800 Ci/MTU
8 Ci/MTU
2.0 Ci/MTU
0.02 Ci/MTU
0.04 Ci/MTU
0.0004 Ci/MTU
0.6-6 Ci/MTU
6xlO~ Ci/MTU
6x10" Ci/MTU
Capital/plant
NA
c
$3x10
6
$3x10
$1.5x10
NA
Unknown
NA
c
$1.2x10
NA. -
c
$1.2x10
NA
$1.0xl05
$3. 5x10 5
Operation/yr
NA
R
$1x10
5
$1.5x10
$1x10
NA w
Unknown 1
tn
NA
Unknown
NA
Unknown
NA
%
$5.0x10
$7.0xlo'*
Notes:
1) NA indicates not applicable
2) Iodine-131 content is estimated for fuel with 33,000
MWd/MTU burnup and 150 day cooling
-------
B-16
The processes have been previously reviewed by several authors
(lef. B-18, B-19, B-20, B-21, and B-22) with regard to development
status, advantages and disadvantages, cost, and efficiency. Kirk
(Ref. B-18) has prepared a comprehensive review of the radiation
hazard from krypton-85.. The conclusions reached in these reviews
indicate that at present only cryogenic distillation, selective
absorption processes, and eryegenic adsorption are worthy of
consideration for control of krypton-85 discharges from reprocessing
plants. (See table B.3 from Ref. B-19.) Systems based upon both the
cryogenic adsorption and cryogenic distillation processes have been
designed for and are being installed at light-water-reactors for
extension of holdup times for gaseous effluents containing noble gas
radionuclides. The selective absorption process has been developed
for application to reactor systems, but requires further development
to be applicable to fuel reprocessing plants (Ref. B-20).
The cost of krypton collection systems is highly dependent on the
design of the dissolution process. Essentially all of the, krypton
present in the spent fuel is released during this dissolving or
leaching process. To minimize the costs, it is necessary to minimize
the total volume of off gases from this process since all of the off
gas must be treated to remove the krypton. Therefore, in a new plant
the cost of a krypton collection system would probably be
significantly less than installation"of such a system in an operating
plant where no effort was'made to minimize the total off gas from the
leaching process. The costs presented are typical of costs for a
-------
TABLE B.3
COMPARISON OF PROCESSES FOR THE REMOVAL OF 85Kr FROM DtSSOLVER OFF-GAS
A FUEL-REPROCESSING PLAIT '
Process
1. Room -temperature
charcoal beds or
molecular sieves
2. Low-temperature
cKsrcoal beds or
silica gel
3. Cryogenic
distillation
4. Liquid extraction
5. Clathr ate precipitation
Kryptonates
6. Perm-selective
membranes
7. Thermal diffusion
8. Electrostatic
diffusion
Kr
recovery
Oft)
• 99
99
98
99
Unknown
99
Development
status
Bench scale completed;
scale -up feasible.
Development completed;
plant operated
Developed and
operated en a
significant scale
Bench scale completed;
demonstration needed
for large scale
Laboratory studies
only? no engineering
Bench-scale worki
need engineering tests
Little pertinent data
Limited; technical
feasibility not proven
Advantages
Simple operation;
accepts dilute feed
gas
Small-volume beds;
uses dilute feed
gas
Low capital cost and
low operating cost
Using Freon-12: low -
refrigeration costs;
low solvent costs;
no explosion hazard;
might eliminate
pie -treatment
ssKr is collected
as a solid;storable
Disadvantages
Large-volume adsorber beds;
charcoal can ignite; . strong
oxidizing gases must be removed
prior to adsorption
Charcoal can ignite; oxidizing
gases, CO2 and tL^O must be
removed! large consumption
of liquid nitrogen: adsorbers
must withstand high pressure;
high operating cost.
Explosion hazard in forming and
concentrating ozone
The absorber column operates at
200 lb/in2 (gauge); the volume of
extractant is large if operated
at 15 Ib/in2 (gauge)
Meeds concentrated feed gasj
crystallization step slow
Membranes sensitive to chemicals;
high power costs
Poor economics for disposal
of dilute 85Kr waste
This Table is reprinted with the kind permission of Dr. C. M. Slansky who presented it in
a paper published in Atomic Energy Review CSee reference B-19)
-------
B-18
krypton control system In a new facility. At the Midwest facility an
effort was made to prevent large scale dilution of the off gas.
Cryogenic adsorption systems remove krypton, .from process gas
streams by adsorption in refrigerated activated charcoal beds until
the bed capacity is reached, followed by desorption into a purging
gas while heating the beds. This was demonstrated on a large scale
at the Idaho Chemical Processing Plant (ICPP) more than 15 years ago.
The disadvantages of this process are high refrigeration costs, fire
hazard potential, explosion potential due to hydrocarbons, nitrogen
oxides and ozone, and impurity plugging of the adsorbers. An overall
recovery fraction of 311 was obtained for krypton although with
design changes and modifications in operating procedures, a recovery
fraction of the order of 99% could be achieved. The system can be
considered but may not be the best for application to fuel
reprocessing plants. It may have more potential for application to
interim holdup for effluent gases in reactors.
Experience has been gained for this method in the development of
the HTG1 and a decontamination factor of 10 appears to be achievable
under most operating conditions. Operation of the adsorber beds at ,
cryogenic temperatures helps overcome the problem of an occasional
abrupt release of adsorbed contaminants which has been experienced
with ambient temperature absorbed beds. Assigning a decontamination
factor to ambient temperature systems is questionable since
experience has shown that under adverse conditions it is possible to
experience a negative decontamination factor for interim periods.
-------
B-19
Total capital costs, including installation, for a cryogenic
adsorption system are estimated at 3 million dollars based on general
estimates for use of systems at reactors. Slansky (Ref. B-19)
estimated one million dollars for capital costs which appears low.
An annual operating cost of $150,000 is reasonable.
Cryogenic distillation, which is based upon separation of gases
due to differences in their relative volatilities at low
temperatures, has been demonstrated and operated on a significant
scale at ICPP for removal of krypton and xenon from an off-gas stream
(Ref. B-23 and B-24). Recovery of krypton and xenon in a form
suitable for bottling in gas cylinders is possible with this process.
An additional advantage is the lower capital and operating cost of
this cryogenic system as compared to cryogenic adsorption. The
cryogenic distillation process entails some potential for explosion.
In spite of the explosion potential, cryogenic distillation is
considered to be one of the two most promising processes for noble
gas control at reprocessing plants.
Considerable experience has been gained in operating these
systems in liquified gas (or air) plants. While decontamination
3
factors across the cryogenic stage of 10 have been estimated, small
leaks can occur in a system. Estimates of the recovery factor for an
overall system range from 98% to 99.99%. Decontamination factors of
3
10 (99.9% recovery) should be attainable with this system and
guarantees of such performance have been submitted to the General
Electric Co, with bids for installation of a cryogenic distillation
system at their Midwest plant (Ref. B-26),
-------
B-20
The General Electric Co. received bids ranging from 0.75 to 1.5
million dollars for the equipment needed to install a cryogenic
distillation system in their Midwest plant (lef. B-26). It is
estimated that installation costs would match equipment costs, thus
producing a total capital cost of about 3 million dollars for this
system. Annual operating costs of $100,000, postulated by Slansky
(Ref. B-19)» are probably a good estimate. Xenon is also recovered
by this system and if kept separate from the krypton, has a potential
market value.
Selective absorption (liquid extraction) depends upon the
relative solubilities of gases in the solvent used — Freon-12 being
the typical solvent under consideration. Krypton and xenon are
selectively absorbed in this solvent while other materials pass
through. The solvent is then processed to recover the krypton and
xenon which can be stored while the solvent is recycled. Bench-scale
studies have been completed (Ref. B-20 and B-25) and, as indicated
previously, commercial systems for application to reactors have been
developed. Because of the anticipated explosion hazard associated
with the cryogenic systems, management personnel of commercial fuel
reprocessing facilities (NFS, Barnwell) have indicated a preference
for fluorocarbon selective absorption systems of this type.
Major disadvantages of such a system include radiation
degradation of the solvent, and a requirement for pretreatment of gas
streams to increase system tolerance to impurities. Babcock and
Wilcox indicates that these problems can be circumvented with good
-------
B-21
engineering design. Decontamination factors of 100 to 1000 are
expected for such systems (Ref. B-20).
The cost of a selective absorption system for installation at a
reactor is estimated at one million dollars. Assuming a slightly
greater capacity system for a reprocessing plant results in a total
installed cost of $1.5 million. Annual operating costs of $100,000
predicted by Slansky (Ref. B-19) also appear reasonable.
VIII. ' TRITIUM.CONTROL
Processes available for tritium control have recently been
reviewed (Ref. B-27). The techniques considered include: chemical
exchange, distillation, electrolysis, diffusion and centrifugation,
radiolysis, adsorption and chromography, solvent extraction, and
molecular excitation. At present the information on these techniques
is inadequate to project the technical or economic feasibility of
retaining tritium.
Voloxidation is a new method of tritium control that looks
promising (Ref. B-ll), In this process the fuel pins, just after
shearing, are heated to approximately 650° C in a stream of air or
oxygen. Tritiated water is generated which should be relatively free
of ordinary water and consequently occupy a much smaller volume than
tritium wastes from presently planned 6r operating reprocessing
plants. This process may collect approximately 99% of the tritium
which is present in the unprocessed fuel. Since the voloxidation
process requires a major change in the head-end design of fuel
reprocessing plants, it would probably be impractical to back-fit
existing or planned facilities.
-------
B-22
The projected release rate of tritium•from a facility having no
tritium control systems is 88 Ci/MTO. Releases may be either by the
air or water pathways depending on plant design. With the
voloxidation head-end process, tritium decontamination factors of 100
are anticipated. Cost information on the voloxidation process is not
presently available5 however, General Electric considered three other
methods of tritium recovery and/or disposal in connection with their
Midwest plant (Ref. B-26). Removal from off gas was stated to cost
approximately $10 million with no process or technical feasibility
defined. Deep-well disposal of tritiated water was reported to run
between $400,000 and $500.000. Finally, off site shipment cost was
estimated to be between $250,000 and $350,000 with no estimate of
feasibility or safety associated with the packaging and transport,
IX. RADIOIOPIN1 CONTROL
A variety of processes have been developed and used for
collection of radioactive iodine. These processes include (1) wet
collection — aqueous scrubbing (reactive sprays, towers, wet
filters); (2) adsorption (charcoal, activated charcoal, silver
zeolite metallic filters); and (3) filtration (high efficiency
particulate — HEPA,, sand deep-bed fiber glass).
The particular treatment process to be selected depends upon the
collection efficiency desired, quantity and chemical species of
iodine involved, stream characteristics, i.e., flow rates,
temperature, humidity, etc. The actual overall removal efficiency of
a system is dependent upon the off-gas flow paths as well as the
characteristics of the treatment techniques.
-------
B-23
WetCollection Techniques have been used to quantitatively retain
iodine in the liquid phase by adding mercury salts during fuel
dissolution at the Savannah River Plant. The resulting Hg-I complex
is solvent extracted and the solvent is washed in a solvent scrubber
to remove any remaining iodine. Disposal of the solid Hg-I complex
must be effected. However, there are no AEC guidelines for disposal
of such waste at present.
A mercuric iodate precipitate can result from scrubbing the off
gas with a mixture of 8 to 14 molar nitric acid and 0.2 to 0.4 molar
Hg(NO,,),j (mercuric nitrate). Solid iodine can be obtained by use of
concentrated 17 to 19 molar nitric acid at room temperature. The
latter systems can be considered capable of removing massive
concentrations of elemental iodine and trace quantities of other
k
forms. Decontamination factors of 10 for all forms of iodine are
reputed to be theoretically feasible in either packed or bubble-
packed scrubbers at operating throughput rates considered appropriate
for fuel reprocessing (Ref. B-28). If the off gas contains high
concentrations of NO some loss in the decontamination factor occurs.
Alkaline solutions (NaOH, NaHC03, NaC03) are also used in scrubber
columns and have reported decontamination factors of 10 to 20, In
packed columns, the separation efficiency is a function of bed
height, fiber drag coefficient, fiber diameter and fiber volume
fraction.
An operational facility (Euroehemi Fuel Reprocessing Plant) which
uses two scrubbers (NaOH for low iodine concentrations and NaHCO or
w
Hg(NOg) for high iodine concentrations) reports decontamination
-------
B-24
factors of 500 for gaseous iodine and 2,500 for iodine in aerosols.
The chemical form of iodine which emanates from these scrubbers is
predominantly organic (Ref. B-29) although traces of hypoiodous acid
have also been reported.
Reactive sprays, hydrazine and thiosulfate, have been studied for
application to iodine and methyl iodide washout from reactor
containment atmospheres with decontamination factors of 2,000 and 100
reported respectively. Such systems have not been applied to fuel
reprocessing plants. Their performance is found to be a function of
relative humidity, temperature, drop size, solution pH, and
concentration,
Charcoal Adsorption has been used for more than 10 years in
reactor and fuel reprocessing off-gas systems. The iodine removal
efficiencies of activated charcoal that have been reported cover a
broad range (50-99.99%). The actual decontamination factor is
dependent upon the forms of iodine, concentration, stream humidity
and flow velocity, and charcoal impregnant.
Silver zeolite (AgN03 impregnated in an alumina-silica molecular
sieve) is reputed to be superior to impregnated charcoal for the
adsorption of methyl iodide (approximately 20 times that for
impregnated charcoal under dry air conditions)(Eef. B-30). The
principal design consideration is the effective residence time in the
bed which is related to the face velocity. If the silver zeolite
(AgZ) beds are designed to permit a mean residence time of about 0.5
seconds, efficiencies are greater than 99.91 under the most adverse
conditions of humidity and chemical form of the iodine (Ref. B-30 and
-------
B-25
B-31), Eemoval efficiencies can be optimized through selection of
the zeolite material with its characteristic alumina-silica ratio.
However, this ratio largely determines the resistance of the AgZ to
acid vapors and thus the effective life of the bed. In general, high
acid resistance results in lower removal efficiencies.
Although the efficiency, of the AgZ is acceptable, consideration
must be given to the loading characteristics of AgZ in terms of the
total iodine cleanup system. The AgZ beds will adsorb all halogens
and probably cannot be used when HC1 is used for fuel dissolution.
In addition, there is a considerable amount of iodine (iodine-129 and
-127) present in the off gas, and this may load the AgZ at
unacceptably high rates. Current technology employs the alkaline
scrubber to remove a large fraction of this iodine which is
theoretically elemental in composition. Work is continuing on the
development of other metallic zeolites which may eventually replace
the scrubbers. Lead zeolite currently appears to be the most
attractive of these more economical and less efficient systems.
The lifetime of the AgZ system is improved by introduction of an
oxidizing catalyst upstream of the sorbent (Ref. B-28), The use of
AgZ systems has been shown to be feasible and practicable, but full
scale operation has yet to be accomplished. Adsorption of iodine on
metallic filters (copper, steel or aluminum, silver-coated copper)
has resulted in recovery in the range of 97.4 to 99.9%. Silver-
coated silica gel has a reported decontamination factor of 10 for
iodine (elemental).
-------
B-26
Filtration is primarily designed for control of participates. It
cannot be seriously considered as a primary technique for iodine
collection, since it depends upon sorption of iodine to particulates
which are then trapped by the filter. In any event, all off-gas
streams from fuel reprocessing plants will be filtered through sand,
deep-bed fiber glass or high efficiency particulate air filters in
addition to the iodine control systems. The iodine collected by
particulate filters will experience desorption at a rate which is
dependent on stream conditions. While this desorption rate may be
significant for the short-lived iodine-131 discharge, it will have a
negligible effect on the iodine-129 discharge. Therefore, the
removal efficiency for filtration of iodine is negligible,
Control Systems Evaluation. It is estimated (Ref. B-32) that
much of the discharged iodine will be in an organic form and that
most of this iodine will interact with atmospheric particles and
settle out within 10 to 20 miles of the discharge point. Thus, the
discharged iodine appears to be principally a local problem, although
long-term environmental transport must be considered because of
iodine-129 with its long half-life.
LWR spent fuel with a burnup of 33,000 megawatt days per metric
ton will contain about 2 Ci of iodine~131 per metric ton after 150
days cooling. In addition to being dependent on the irradiation .
history of the spent fuel, the iodine-131 content decreases rapidly
with the cooling period because of its 8 day half-life. The iodine-
129 content will be about 0.04 Ci per metric ton of spent fuel.
-------
B-27
An overall system for iodine collection in fuel reprocessing
plants will probably consist of a combination of wet collectors
(liquids and off-gas scrubbers) followed by a catalytic decomposition
system, an adsorption system, and a filtration system. The overall
efficiency will be dependent upon the detailed design of the off-gas
flow system. For example, if the system is assembled in the manner
of the Midwest facility, the off-gas stream bearing the iodine is
first processed through the caustic scrubber (eff. - 90%) and then
through the AgZ bed. The gas stream is then routed to the sand
filter and stack. The overall removal efficiency for this treatment
should be greater than 99.9%. However, the scrubber solution,
theoretically containing 90% of the iodine, is routed to the low-
level-waste storage vault. The low-level waste in the vault is
routinely evaporated for volume reduction, opening a way for a large
fraction of the iodine to be revolatilized. This second off-gas
stream is processed through.a second scrubber but no AgZ bed.
Therefore, about 10% of the iodine routed to the low-level vault can
be discharged to the sand filter and the stack. The buildup of the
long-lived iodine-129 in the waste storage vault must also be
considered in this system since it is reasonable to assume that the
iodine-129 discharge will increase with the inventory buildup.
The development work previously discussed may produce more
effective iodine cleaning systems for reprocessing plant off gases
(lef. B-ll, B-12, and B-13). However, even with present technology
it is difficult to speculate what the cleaning efficiencies of actual
installed systems will be. Once these systems are in operation their
-------
B-28
performance can be monitored and documented. In the interim the use
of an overall decontamination factbr of 100 to 1,000 appears
acceptable.
Based on experimental evidence it is reasonable to assume that
the wet scrubber/AgS system can be^designed to achieve a minimum
decontamination factor of 1,000 under most conditions, However, as
discussed previously, the performance of the off~gas iodine cleaning
system may not be the controlling factor in determining total iodine
discharges from a reprocessing plant. In particular, the system for
handling or processing the scrubber solutions must minimally provide
the same cleaning efficiency as the off-gas stream since a large
fraction of the iodine is expected to be in the scrubber solutions.
Therefore^ it can be concluded that while the efficiency of currently
planned iodine removal systems will be a minimum of 99,9%, the total
iodine waste handling system introduces a large uncertainty.
Meager data are available regarding the costs for iodine removal
systems. The costs are dependent upon the volume of off gas which
requires processing as well as the total quantity of iodine to be
removed. The following estimates have been adjusted to 1970 dollars
using the Marshall and Stevens Equipment Index.
-------
B-29
Wet Collection Capital Costs* ••• Operating Costs*
a. Spray tower $310-620/1,000 cfm Unknown
b. Packed towers $620-1,240/1,000 cfm $510-1,020/1,000 cfm
c. Solvent extraction Unknown Unknown .
Absorbing Systems: • ' •• ' ' '
a. Charcoal $40,000/bed Unknown
b. Charcoal + catalytic
combiner (KRB) $750,000^$10 Unknown
c. Ag zeolite $400,QQO/bed Unknown
Complete SystemEstimate;
5 MTU/day throughput g k
a. Caustic scrubbers $7.5 x 10 $7.5. x 10
* At an average plant the total air volume moved is approximately
60,000 cfm.
The costs listed above are for the caustic Scrubber systems only
since the quantities of AgZ needed will not be known until operating
experience is gained to determine the extent, of the .parasitic loading.
-------
B-30
the alternative of using holdup for control of iodine-131 from fuel
reprocessing plants is applicable only in conjunction with a removal
system that will extract the long-lived radioiodine, iodine-129.
Estimates of the cost involved for holdup are as follows:
Given: plutonium price $7/g
annual interest rate 10%
LWR fuel plutonium content 16* kg/MTU
LMFBR fuel plutonium content 86 kg/MT
time increment (1 month) 30 days
MlES. x II x MP. x i-2SSE . $930/tonne month
MOT g year 12 month
Example: Barnwell will have a capacity of 1,500 MTU/year and will
process at 150 days cooling time. If the cooling time is increased to
6
300 days to permit the iodine-131 to decay by a factor of about 10 , the
cost will be:
930 $ 1.500 MTU .$6.975.000
MTU month x year x 5 »°at*8 " year
For each additional 30 days of holdup the cost would be:
930 $ , x 1,500-mu x l month m $1,395,000
MTU month year year
For LMFBR fuel cooled for 150 days, the cost would be approximately
10 times greater.
*Based on 2/3 UO,, fuel and 1/3 mixed oxide fuel. (See table A-4)
-------
B-31
X. PMTICUIATE CONTROL •
Radioactive particulates associated with fuel reprocessing will
be found in both the liquid and the gaseous effluent streams. The
particulates arise due to the various operations (shearing through
waste solidification) and as a consequence can include a variety of
radioisotopes. They will also be in a variety of chemical and
physical forms. Easily condensable vapors should be considered as a
source of particulates bearing radioactivity. Most of the
radioactive wastes available for discharges are attached to
particulates. Included in this category are the isotopes in the
actinide series, many of which are highly radiotoxie since they decay
by alpha emission. Other radionuclides which potentially can
contribute significantly to the waste gas stream particulate makeup
include the volatile isotopes, such as ruthenium-106, cesium-134, and
the longer-lived isotopes, such as strontium-90 and eesium-137.
The particulate effluent from reprocessing plants will for the
most part be soluble and will quite probably be in the nitrate form.
The particulate effluents from the solidification process will
probably be oxides and thus insoluble. Ruthenium may be an
exception, especially from the solidification process, and may be
complexed as an organic.
Probably the most acceptable theoretical estimates of discharged
radionuclides were made by Oak Ridge scientists (Ref. B-8). These
estimates appear conservative, however, as indicated by the gaseous
discharges measured at the NFS facility (Ref. 1-17). An annual
discharge of 1.0 to 10 Ci of beta-gamma fission products, excluding
-------
B-32
tritium, noble gases and the iodines, currently is the most
reasonable estimate. The routine annual gaseous discharge of
actinides at the DSAEC Rocky Flats Plutonium Recovery Facility (Ref.
B-37) averaged; 2i4 millicuries from 1953 to 1970. Fuel reprocessing
plants may be expected to have higher discharges (see table B.2).
Control Techniques - Particulars are generally collected from
gaseous waste streams through the use of inertial separators
(cyclone, or gravity settling), filtration (fabric, glassfil,
sandbeds, HEPA), precipitation (electric, thermal) sonic
agglomeration, and liquid scrubbing. The specific system used is
dependent upon the type of source and efficiency of control desired.
'.Filtration - Because of the concern regarding the inhalation of
••• . '•>. •' '. • '•''.''
radioactive particulate material, high efficiency filtration systems
for removal of particulates from air have received the primary
-.-'.' " i
emphasis in the nuclear industry and are widely used (Ref. B-32).
farticulate filters and materials of a variety of types are
available. The materials include cellulose-asbestos, glass, glass-
asbestos , plastic fibers and ceramics. Filters are generally
classified as panel (viscous impingement) filters which are coated
with a tacky substance to increase particle adherence, and extended
medium dry-type filters which are called "bag" or "sock" filters.
Panel filters are designated Group I and have a low efficiency for
small particulates. Bag filters are designated as Group II and Group
III (medium and high efficiency, respectively). HEPA (High
Efficiency Particulate Air) filters are classified as a special
group. Sand filters (Ref. B-33) in general are designed for a
-------
B-33
specific application, usually for use in cleaning corrosive materials
'.'.' •".-('• ' ' • ' ' '
from air streams. Their use in the nuclear field has been limited,
but is expected to increase since they can be considered relatively
fail-safe. Deep-bed sand filters have been employed at Hanford and
Savannah River for reprocessing plant gaseous effluent filtration.
The Midwest facility has also installed a deep-bed sand filter fo.r
final filtration.
The parameters affecting filter performance, in addition to face
velocity characteristics are; resistance, geometrical size, pressure
drop, penetration, collection efficiency, dust and holding capacity,
particle sizes, loading, humidity, temperature, and chemical
resistance.
Liquid Scrubbing - This method has been used in cleaning the off-
gas stream from solidification processes (Eef. B-34 and B~35),
Scrubbers are used as initial cleaning devices and the scrubber
solutions can be recycled to the process streams. Experience is
limited to the Idaho Chemical Processing Plant and the prototype work
conducted under the Waste Solidification Engineering Prototype
program conducted by Battelle at Hanford.
Control Systems Evaluation - The various methods of particulars
control may be grouped in the order of increasing efficiency as
follows:
Fair
Group I panel filters
Good
a) Group II bag filters
-------
B-34
b) Deep-bed sand or fiberglass filters
c) Scrubbers
Better
Group III bag filters
Best
HEPA filters
The anticipated performance of fibrous filters can vary from 2 to
99.97% retention for 0.3 micrometer particulates and up to 100% for
10 micrometer size. HEPA filters have consistently demonstrated
penetrations less than 0,3xlO~3 (99.97% efficiency). However, the
overall system efficiency is highly dependent on filter integrity,
proper installation, operating conditions (particularly air flow
rates and loading), and aging characteristics. Adequate in-place
testing is necessary since improper installation or filter damage in
shipment or installation could result in leakages up to 30%.
A typical filter installation consists of a prefilter (roughing
filter) followed by two HEPA filters in series or by a HEPA filter
and a sand filter. The prefilter generally has a rating of greater
than 75% (Group III for 1.0 micrometer particles) and offers a
considerable cost savings by reducing mass loadings on the more
costly HEPA filters. The HEPA filters are rated at a minimum
efficiency of 99.97% for 0,3 micrometer particulates. The reported
efficiency of deep-bed sand filters for submicrometer particles is
greater than 99%. Thus, the overall decontamination factor for the
5
filter themselves should be on the order of 10 . However, the
efficiency of the system is also dependent upon factors such as leak
-------
B-35
tightness and aging characteristics. The minimally acceptable in-
place test results would probably be in the order of 99.97% for a
typical HEPA system (it is emphasized this is a system test). The
system would probably be assigned a credit of 99% to 99.9%, the
actual credit depending on the testing frequency.
Capital and operating costs are highly dependent on volume flow
rates. Based on Silverman's estimates (Ref. B-36), a HEPA filter
would have a capital cost between $200-800/1000 cfm and annual
operating costs between $100-450/1000 cfm depending upon
specifications, corrosion resistance, and other operating conditions.
A pre-filter for reducing the HEPA loading would cost between $100-
300/1000 cfm for capital cost and $15-25/1000 cfm for annual
operation. Using approximate costs and a total flow of 60,000 cfm
the system costs would be:
HEPA Pre-filter
Capital cost (X60) $48,000 $12,000
Annual operating costs (X60) $18,000 $ 1,200
40-year operating costs
(present worth (§10%) . $176,000 $11,700
Total costs (1960 prices) $224,000 $23,700
Conversion to 1970 price -
(Miller-Stevens index 1.28) $287,000 $30,400
The costs of a glass-fiber or sand deep-bed filters were reported
by Silverman (Ref. B-36) to be between $2,860-5,500 capital costs per
100 cfm with associated operating cost between $400 and $800 per 100
cfm (based on only depreciation and air-moving costs). Taking a
representative tidal flow of 60,000 cfm the cost of the deep-bed
system would be:
-------
B-36
Capital cost ($3,500 x 60) $210,000
Annual operating cost ($800 x60) $ 48,000
40-year operating costs
(present worth @10% discount) $470,000
Total cost (1960 prices) $680,000
Conversion to 1970 prices Total $870,000
The reported capital cost for the sand-bed filter, discharge
stack, and associated equipment is $400,000 for the Midwest facility
(Ref. B-33). The corrected capital cost to 1970 for. a deep-bed sand
filter is $269,000 which agrees well with the estimate for Midwest.
The total filter installation and 40-year operating costs at
reprocessing plants are:
1970 Present Worth (§10%
Prefilter + filter +
Deep-bed sand filter $1,187,000
Prefilter + HEPA filter +
HEPA filter $ 604,000
-------
B-37
REFERENCES
B-l. Haas, W. 0., Jr., Solvent Extraction: General Principles,
Chapter 4, Chemical Processing of Reactor Fuels, Flagg, ed.,
Academic Press, New York, 1961.
B-2. Siddall, T. H., Ill, Solvent Extraction Processes Based on
Tri-n-butyl Phosphate, Chapter V, ChemicalProcessing of
Reactor Fuels, Flagg, ed., Academic Press, New York, 1961.
B-3. Benedict, M., and T. H. Pigford, Nuclear Chemical Engineering,
McGraw-Hill, New fork, 1957.
B-4. Stoller, S. M., and R. B. Richards, eds., Reactor Handbook
Volume II Fuel Reprocessing., 2nd Edition, Interscience,
New York, 1961.
B-5. Goode, J. H., Hot Cell Evaluation of the Release of Tritium and
85Krypton During Processing of ThO.-UQ Fuels, ORNL-3956,
June 1966.
B-6. Cochran, J. A., et. al., An Investigation of Airborne Radio-
Active Effluent from an Operating Nuclear Fuel Reprocessing
Plant, BRH/NERHL 70-3, July 1970.
B-7. Cochran, J. A., W. R. Griffin, Jr., and E. J. Troianello,
Characterization of Tritium Stack Effluent from a Fuel
Reprocessing Plant, ANS Transactions, 15, 1, Las Vegas,
June 1972.
B-8, Oak Ridge National Laboratory, Siting of Fuel Reprocessing
Plants and Waste Management Facilities, ORNL-4451, July 1970.
B-9. Commander, R. E., et. al.» Operation of the Waste Calcining Facility
with Highly Radioactive Aqueous Waste, ICPP, IDO-14662, June 1966.
B-10. Magno. P. J., et. al., Liquid Waste Effluents from a Nuclear Fuel
Reprocessing Plant, BRH/N1RHL 70-2, November 1970.
B-ll. Oak Ridge National Laboratory, Chemical Technology Annual Progress
Report, ORNL-4794.
B-12. Argonne National Laboratory, Reactor Development Program Progress
Report, ANL-7872, October 1971.
B-13. Krumpelt, M., et. a., The Containment of Fission Product Iodine in
the Reprocessing of LMFBR Fuels by Pyrochemical Reactions, Nuclear
Technology, 15, September 1972.
-------
B-38
B-14. Oak Ridge National Laboratory, Aqueous Processing of LMFBR Fuels
Technical Assessment and Experimental Program Definition, ORNL-
4436, June 1970
B-15. Oak Ridge National Laboratory, National HTGR Fuel Recycle
Development Program Plan, ORNL-4702, August 1971.
B-16. Code of Federal Regulations, Title 10, Part 50, Appendix F,
Policy Relating to the Siting of Fuel Reprocessing Plants and
Related Waste Management Facilities, November 1970,
B-17, Russell, J. L. and F. L. Galpin, A Review of.Measured and
Estimated Offsite Doses at Fuel Reprocessing Plants, NEA/IAEA
Syriposium on the Management of Radioactive Wastes from Fuel
Reprocessing, Paris, November 1972.
B-18. Kirk, W. P., 85Krypton - A Review of the Literature and
Analysis of Radiation Hazards, EPA, ORP, Washington, January 1972.
B-19. Slansky, C. M., Separation Processes for Noble Gas Fission
Products from the Off-Gas of Fuel Reprocessing Plants, Atomic
EnergyReview, 9, 2, IAEA, Vienna.
B-20. Merriman, H. R., M. J. Stephenson, and D. I. Dunthorn, Recent
Developments in Controlling the Release of Noble Gases by
Absorption in Fluorocarbons, ANS Transactions, 15s 1, Las Vegas,
1972.
B-21. Keilholtz, G, W., Krypton-Xenon Removal Systems, Nuclear Safety
12, 6, November-December 1971.
B-22. Nichols, J. P. and F. T. Binford, Status of Noble Gas Removal
and Disposal ORNL-TM-3515, August 1971 ,
B-23. Bendixsen, C, K., and G. F. Offutt, Rare Gas Recovery Facility
at the Idaho Chemical Processing Plant, Idaho Nuclear Corp.,
IN-1221, April 1969.
B-24. Benixsen C, L., and F. 0. German, Operation of the ICPP Rare
Gas Recovery Facility During Fiscal Year 1970, Allied Chemical
Corp., ICP-1001, October 1971.
B-25. Stephenson, M. J. J. R. Merriman, and D. I. Dunthorn, Experimental
Investigation of the Removal of Krypton and Xenon from Contaminated
Gas Streams by Selective Absorption in Fluorocarbon Solvents:
Phase I Completion Report, Union Carbide, K-1780, August 1970.
-------
B-39
B-26. General Electric Company, Midwest Fuel Recovery Plant,
Applicant's Environmental Report, Supplement 1, NED 14504-2
November 1971 -
B-27. Environmental Protection Agency, The Separation and Control of
Tritium State-of~the-Art Study, Pacific Northwest Laboratories.
BMI, April 1972.
B-28. ORNL-4794 Chem. Tech. Div,, Annual Progress Report Period
Ending March 31, 1972.
B-29. Personal Communication, 0. 0. Yarboro, Oak Ridge National
Laboratory Chemical Technology Division ,
B-30. Pence, D. T., P. A. Duce, and W. J. Maeck, Application of Metal
Zeolites to Nuclear Fuel Reprocessing Plant Off-Gas Treatment,
ANS Transactions, 15, 1, Las Vegas, 1972.
B-31. Wilhelm, J. G., and H. Schuttelkopf, An Inorganic Adsorber
Material for Off-Gas Cleaning in Fuel Reprocessing Plants,
12th AEG Air Cleaning Conference, Proceedings to be published
August 1972.
B-32. Burchsted, C. A., and A. B. Fuller, Design Construction, and Testing
of High Efficiency Air Filtration Systems for Nuclear Application,
ORNL-NSIC-65, January 1970.
B-33. Juvinall, R. A., R. W. Kessil, and M. J. Steindler, Sand-Bed
Filtration of Aerosols; A Review of Published Information on
Their Use in Industrial and Stomic Energy Facilities. Argonne
National Laboratory, ANL-7683, June 1970.
B-34. Lakey, L. T. and B. R. Wheeler, Solidification of High-Level
Radioactive Wastes at the Idaho Chemical Processing Plant,
to be presented at NEA/IAEA Symposium on the Management of
Radioactive Wastes from Fuel Reprocessing, November 1972.
B-35. McElroy, J. L., A. G. Blaservits, and 1. J. Schneider, Status
of the Waste Solidification Demonstration Program, Nuclear
Technology, 12, September 1971.
B-36. Silvernan, L., Economic Aspects of Air and Gas Cleaning for
Nuclear Energy Processes, Proceeding of a Conference on the
Disposal of Radioactive Wastes. Vol. I, IAEA, Vienna, 1960.
B-37. U.S. Atomic Energy Commission, Environmental Statement
Plutonium Recovery Facility Rocky Flats Plant Colorado,
WASH-1507, January 1972.
-------
B-40
B-38. Report of the Committee to Study the Establishment of Plants
or Facilities for the Recovery of Nuclear Fuel and the
Storage of Waste Nuclear Materials, 1972.
B-39, Hearings on Midwest Plant before Illinois Pollution Control
Board.
B-40, Martin, J. A., Jr., Calculations of Doses in 1971 Due
to Radionuclides Emitted by Nuclear Fuel Services Fuel
Reprocessing Plant, Radiat Data Rep. 14:59-76 (February 1973)-
-------
APPENDIX C
RADIOLOGICAL DOSE AND
HEALTH IMPACT CONVERSION FACTORS
-------
-------
C-l
I. INTRODUCTION
Radioactive materials released into the environment from nuclear
fuel reprocessing become dispersed in the surrounding media (air, water,
etc.) and ultimately may produce health effects in man. The impact
of a given radionuclide release on the population surrounding a source
is assessed here in terms of three factors: (1) a dilution factor to
calculate the concentration of the released activity in the medium of
interest, (2) a medium concentration to dose conversion factor, and
(3) a risk factor which relates the likelihood of a given biological
effect to an absorbed dose of one rad. 1 These factors are discussed
below for tritium, krypton-85, the radioiodines, and alpha-emitting
transuranics (such as plutonium).
II. PATHWAYS
Releases of radionuclides from a nuclear fuel reprocessing plant
can occur by venting through an exhaust stack to the atmosphere or
drainage to a nearby waterway. The principal pathway of concern in
assessing the health impact due to nuclear fuel reprocessing plant
operations is the atmospheric pathway because such releases can
become dispersed in any direction and lead directly to radiation
exposure to man. The significance of the water pathway is expected to
be quite small for future reprocessing plants because presently proposed
designs do not plan on any releases to waterways.
Atmospheric dispersion of radioactivity has been discussed by a
number of authors (Ref. C-l, C-2) and for the case of interest here, a
Rad - The unit of energy imparted to matter by ionizing radiation
and equal to ,01 J/kg in any medium.
-------
C-2
gaussian plume diffusion model is usually assumed to be the best choice.
At distances relatively close to a source (2-3 km) this model can
predict the air concentration of krypton-85, for example, within a
factor of two or three. Its applicability at longer distances depends
upon the local weather conditions at the time of radioactive release
and the topography. For unstable atmospheric conditions, it may be
reasonably accurate as far as 10 km from a source. Since the average
radionuclide concentrations in air around sites have been calculated
for distances as far as 80 km (SO miles) from the source, it is obvious
that the validity of the atmospheric transport model used is an important
limitation. However, the point here is to examine the general case
and provide an overall index of health risk; the risk from a particular
plant will depend on the details of the local meteorological situation.
For worldwide distribution of gases, uniform dispersion was assumed in
determining air concentrations.
An important radionuclide pathway for man is the direct contam-
ination of foodstuffs - particularly milk. For iodine-131, this
pathway has been studied extensively by several authors including
Garner (Ref. C-3) and Bryant (Eef. C-5, C-5). Long-term buildup of
the isotope iodine-129 may be important due to its half-life of
17xl06 years but appreciable buildup has not been documented
in the literature. The short-term behavior of iodine-129 has been
considered, however (Ref. C-6), and health risk estimates are given
for both iodine-129 and iodine-131.
-------
C-3
III. MEDIA CONCENTRATION TO DOSE CONVERSION FACTORS
Organ or total body dose estimates are critically sensitive to
assumptions concerning the route of uptake, the amount of radioactivity
inhaled or ingested daily, the fraction of activity retained in the- organ
of interest, and the residence -time of the activity in various parts of
the body. The final necessary elements entering into the dose computations
are the physical considerations of organ mass and radionuclide distri-
bution within the organ. In the present state of the art, the complex-
ities of the radionuclide distribution within organs are nearly always
circumvented by assutalng a uniform depositon. Information concerning
the other inputs is based mainly on empirical evidence, largely gathered
from fallout studies and medical investigations. In order to reduce
the number of variables to be considered in dose calculations, the
International Commission on Radiological Protection (ICRP) has postu-
lated a "standard man;" i.e., a model.system having standardized biological
parameters based on either average values or best estimates as listed
in the scientific literature. The standard man is a hypothetical adult
industrial worker and it is not clear to what extent parameters so de-
fined are applicable to an environmentally exposed population.
For particular radionuclides, the sensitivity of certain age groups
may be the limiting factor. In the case of iodine-131, the Federal
Radiation Council (Ref. C-7) has defined children as the most sensitive
population group and, therefore, the biological parameters used in
these media to dose conversion factors are not based on standard man.
Rather, models appropriate for children's thyroid glands and thyroid
metabolism have been used. For the other radionuclides considered here,
-------
C-4
little is known concerning differences between adults and children.
Such differences are seldom considered in the literature. This being
the case, the conversion factors listed in the subsequent sections, while
adequate, must be considered only as first order approximations and
not as definitive dose estimates from environmentally distributed
radionuclides.
Media concentration-to-dose conversion factors are defined below
for krypton-85» tritium, iodine 131, iodine~129, and some of the s
actinides. Other radionuclides are not considered likely to cause
significant environmental exposures of the population based on the
control technologies discussed in appendix B.
A. Rangeof Expected Doses from Krypton-85 Exposure
Since krypton-85 is not metabolized, exposure to humans is limited
to external beta and gamma rays and, to a much lesser extent, the dose
due to krypton-85 dissolved in body fluids. The health risk from
krypton-85 is further limited by the fact that 99% of the decay energy
is dissipated by beta rays which have no potential for deep penetration.
Four target organs are considered for these dose and risk estimates:
total body, gonad, lung, and skin. In each case it can be shown that
only one type of exposure need be considered, the other contributing
an insignificant fraction of the dose.
Kirk (Eef. C-8) has recently reviewed the literature on krypton-85
dose and established relationships between the concentration in air of
krypton-85 and various organ doses. A review of these results show
whidh radiations and source locations are important. For the whole
body, dose and risk estimates can be. based on a consideration of
-------
0-5
external photon exposures, i.e., gamma rays and bremsstrahlung. For
genetic risk calculations the gonadal dose, in the case of males, is
from exposure from external photons; while for females, the whole body
dose estimate can be used. Dose estimates for the lung are based on
internal beta dose plus the total body gamma-ray dose. Skin dose and
risk estimates are based on the dose delivered by external beta
radiation after making an appropriate allowance (0,25) for the shielding
provided by clothing and the nonviable epithelium.
B. Range of Expected Doses from Tritium Exposure
Dose estimates from tritium.exposure have been based on the
assumption that the isotope is contained in body water (Ref. C-9).
However, chronic exposure to environmental tritium has been shown to
lead to incorporation into organic molecules from which tritium is lost
at a slower rate than from body water (Ref. C-10, C-ll). If it is
Assumed that, under equilibrium conditions, all body hydrogen (7.0 kg in
standard man) is uniformly labelled, a sustained concentration of 1 yCi/1
body water would lead to a body burden of 63 uCi, as opposed to 43 yCi if,
as in the ICRP model only, distribution in body water alone is considered.
Ivans (Ref. C-10) found that tritium was not, in fact, uniformly
distributed through deer tissues and, assuming that his observed factors
are applicable to man$ has calculated that the body burden carried by
standard man at a sustained concentration in body water of 1 yCi/1 would
be 60 yCi, i.e., higher by a factor of 1.4 than that based on the ICRP
model. While Evans* factor has been adopted in some dose calculations,
a factor of 1.5 (63/43), although only marginally different, may be a
more appropriate value to use for calculations in man and, therefore, it
is used here.
-------
C-6
Since it is apparent that, under chronic exposure conditions,
tritium may become incorporated into the genetic material (DMA), it has
been suggested that the relative biological effectiveness of tritium
in terms of genetic effects may be greater than unity as a result of DNA
degradation from transmutation and recoil processes in addition to that due
to absorbed energy from ionization processes due to beta emissions. How-
ever, from both experimental (Ref. C-12) and theoretical (lef. C-13)
considerations, it has been concluded that it is the absorbed dose to
mammalian cell nuclei from incorporated internuclear tritium which deter-
mines quantitatively the degree of effect (Ref. C-24). The assumption
made in these calculations that the appropriate value for quality factor
for tritium dose equivalent estimation is 1.0 as recently adopted by the
National Council on Radiation Protection and Measurements (NCRP) (Ref. C-14),
A sustained concentration of 1 yCi tritium per liter of -body water
would thus be equivalent to a specific activity (assuming uniform
labelling of all body hydrogen) of 9x10 3 yCi tritium/g hydrogen, and
would deliver an annual dose to body tissues of approximately 100 mrem.
The concentration of tritium in body water resulting from exposure
to tritium in air is obtained by diluting the daily intake of tritium
by inhalation into the 43 liters of body water with a biological half-
life of 12 days. This amount of tritium is doubled to account for
absorption of tritium through the skin. This leads to an annual 'dose
of 1,7x106 rem2 for an air concentration of 1 Ci tritium/cm^.
the case of beta and gamma rays emitted by fuel reprocessing plant
effluents, the quality factor is one and the dose equivalent in rents
is identical to the dose in rads. Where the effects of such effluents
are considered in this report, doses are expressed in rem units and
biological effects are presented on a per rem basis.
KM - The rem represents that quantity of radiation that is equivalent—
in biological damage of a specified sort - to 1 rad of 250 kVp x rays.
-------
C-7
C. Range of Expected Doses from Iodine-129 and I6difle-131 Exposures
Atmospheric releases of iodine from fuel reprocessing may result
in an accumulation of iodine-129 and iodine-131 in the thyroid glands
of persons living in the surrounding area. The pathway potentially most
hazardous to man for isotopes of iodine is the grass-cow-milk chain,
particularly in cases where the milk is not diluted with uncontaminated
supplies. Direct deposition on foliage will be the only significant
route of contamination of edible herbage for iodlne-131 and is likely to
be the most important for iodine-129, at least over the duration of
plant operation.
Because of the long half-life of iodine-129, recycling through
the soil should be considered. In organically rich soils, the iodine
will be strongly bounded to the soil, but it will be leached rather rapidly
from other types of soils. In any case, plants will incorporate
iodine-129 in ratio proportional to the amount of natural iodine-127
available. The actual amount of iodine-129 incorporated will depend on
the location of the reprocessing plant, and the specific activity of
the iodine-129 (curies of iodine-129 per gram of iodine) in each component
of the terrestrial pathway will change as a function of time as build-
up in the soil increases. At any given time the specific activity in
the ecological chain will be somewhat less that the specific activity of
the iodine-129 in the air. In many cases the specific activity will be
much less because of the large stable iodine reservoir in soils and other
parts of the terrestrial pathway.
-------
C-8
When considering the exposure of individuals to iodine isotopes via
the grass-cow-milk chain, the population potentially at greatest risk
is young children consuming fresh milk (Ref, C-15, C-16). From the data
of Durbin, et al. (Ref. C-1J) the average daily intake of whole cows milk
by U.S. children over the first year of life is about 760 ml. Appropriate
representative data to define the relationship between the amount of
iodine ingested by a 6-month-old child and its concentration in the
critical organ, the thyroid gland, are (Ref. C-6); thyroid weight, 1.8 g;
fraction of ingested iodine reaching critical organ, 0.35; and biolog-
ical half-life of iodine in thyroid, 23 d. Equivalent data for adults,
appropriate to the calculation of average population doses, are: daily
milk consumption, 500 ml (Ref. C-17); thyroid weight, 20 g; fraction
ingested reaching critical organ, 0.3; and biological half-life in
thyroid, 138 d (Ref. C-6). Use of these values yields an annual dose
to the adult thyroid of 0.29 mrem for iodine-131 or 1.9 mrem for iodine-
129 for a concentration of 1 pCi/1 of each isotope. The corresponding
annual doses to the thyroid of children whose daily consumption of milk
during the first year of life contains 1 Ci/1 of the respective
radionuclldes are 4.3 mrem for iodine-131 and 6.3 mrem for iodine-129.
The media conversion factors presented in table C.I are also derived
from considerations discussed in references C-6 and C-7, an assumed
grazing area for a dairy cow of 80 m2 per day and an iodine deposition
velocity of 0.5 cm/s (Ref. C-5).
-------
C-9
TABLE C.I
Milk Concentrations of Iodine-131 and Iodine-129 from
Given Input Concentration and Corresponding Doses
Input Concentration
of Respective Nuclide
Milk Concentrations
Iodine-131
Iodine-129
o
1 pCi/m (ground surface)
1 pCi/m3 (air)
1 pCi/m (air)
1 pCi/m (air)
0.20 pCi/1
6.2xl02 pCi/1
0.28 pCi/1
2.4x103 pCi/1
Annual Dose to Child Thyroid
Iodine-131 Iodine-129
2.7 rem 15 rera
Annual Dose to Adult Thyroid
Iodine-131
Iodine-129
.18 rem
4.6 rem
Estimates of the specific activity (yCi iodine-129/g total iodine) in
the thyroid gland corresponding to an annual dose of 1 rem are, for an
adult, 2.3, and for a 6-month child, 4.1 (lef, C-6). Adoption of a
value of 0.44 rem/yr as the dose delivered to a thyroid gland containing
1 yCi iodine-129/m3 total iodine would thus appear to be a reasonable
estimate for all cases. The mean concentration of stable iodine in the
atmosphere is given as 0.2 Ug/m3 (Ref. C-18). Using this value, it can
0
be shown that an air concentration of 1 pCi iodine-129/m would lead to
an annual thyroid dose of 1.8 rem. In areas where the atmospheric
-------
C-10
concentration of stable iodine-127 is low, the dose could be up to
40 times higher; this upper limit on the dose is set by the amount
of stable iodine-127 released in fuel reprocessing. Thus, for
adults the higher value presented in table C.I of 4.6 rem/yr
per pCi/m3 of iodine-129 in air is selected for use in this study.
D. Range of Expected Doses from Plutonium-239 and Other Actinide Exposures
The potential risks from Inhalation of plutonium-239 depend on whether
the plutonium is in a soluble or an insoluble form. Present experience in-
dicates that, in the case of fuel reprocessing, the plutonium will be pre-
sent in the environment in a relatively insoluble form and the present
dose estimates are based on this assumption. There is also evidence that a
considerable fraction of plutonium-239 inhaled in insoluble form is trans-
located, largely to the bronchial and mediastinal lymph nodes (Ref. C-19).
Since the risk to be pulmonary region depends upon both the amount of
plutonium in the organ and its microdistribution, the region containing
the largest amount of plutonium may not be the region at greatest risk.
Particularly, since the relative sensitivity of the various cell types
encountered has not been established, the dose to the lung from inhaled
particules is calculated on the basis' of an average dose to the entire
pulmonary region for this report. In the case of alpha emitters, such
averaging is obviously inappropriate if there are only a few particles
present. ICRP Publication No. 6 (Ref, C-20) recognizes this and states, "in
the case of the lung, an estimate of the dose equivalent to the critical
tissue determined merely by the product of quality factor and mean dose
may be greatly in error, but further experimental evidence is needed before
a better estimate can be made."
-------
e-ii
In this report, dose estimates are based on the new ICRP lung model
(lef. C-21, C-22), The biological half-life of material in the lung
(pulmonary region) is assumed to be 1,000 days. Using this model,
sustained exposure to an air concentration of 1 pCi/m^ of
plutonium-239 in insoluble form would lead to a dose rate of 12 rem per
year in the pulmonary region. It is assumed that the risk to this
region is representative of the total risk to the lung.
Media-dose conversion factors for other actinide radionuclides
are related to the plutonium-239 conversion factor by taking into
account the effective energy absorbed per disintegration of each
radionuclide and the physical half-life as given in IC1P Publication
Nos. 2 and 6 (Ref C-9, C-20). Table C.2 gives the conversion factors
used in this study for several actinides relative to plutonium-239.
TABLE C.2
Actinide Air-Dose Conversion Factors
Relative to Plutonium-239
Radionuclide
Pu-238
Pu-239
Pu-240
Pu-241
Am-241
Cm-242
Qn-244
a
Relative Conversion Factor
1
1
1
0.001
0.25
0.17
0.33
PI"tonium-239 Conversion Factor - (12xl06 rem/yr)/(l PCi/m3)
-------
C-12
E. Summary
Table C.3 summarizes the media-dose conversion factors presented
in this section. Conversion factors are expressed in terms of
rem/yr resulting from continuous exposure to concentrations
expressed in yCi/cm3 of air.
TABLE C.3
Summary of Media-Dose Conversion Factors
Radionuclide Critical Organ
Kr-85 Whole body
Gonads (female)
Gonads (male)
Lung
Skin
H-3 Whole body
1-129 Infant thyroid
Adult thyroid
1-131 Infant thyroid
Adult thyroid
Pu-239 Lung a
a See paragraph IV D regarding
I?. DOSE-RISK CONVERSION FACTORS
Conversion Factor
(rem/yr) /(yCi/cm3 air)
1.5x10^
1.5x10^
2.0x10^
3.0x10**
50.0x10^
1.7xl06 ot
100 ( rem/yr) /(pCi/em3 water)
15. 0x10 1 2
4.6xl012
2.7xl012
0. 18x10 l 2
12. 0x10 12
consequences of soluble form of plut
Assumptions made in predicting radiation-induced health effects
from nuclear fuel reprocessing are given in this section. Consistent
with recommendations made in the recent (November 1972) National
-------
C-13
Academy of Sciences Committee on Biological Effects of Ionizing
Radiation (BEIR) report (Ref. C-23), the health risks presented in this
report are based on an assumed linear relationship between absorbed
dose and biological effects and that any increased risk is in
addition to that produced by natural radiation; i.e., no threshold
exists. It is further assumed that health effects that have been
observed at dose rates much greater than those likely to be encountered
around fuel reprocessing plants are indicative of radiation effects at
lower dose rates. Only insofar as any biological repair of radiation
damage from low dose rate radiation is neglected do the BEIR
health risk estimates represent upper limits of risk. In most cases
the risk estimates are based on relatively large doses where cell
killing may have influenced the probability of delayed effects being
observed. The BEIR risk estimates used in this report are neither
upper nor lower estimates of risk, but simply the "best available."
As the BEIR report points out, a nonthreshold linear relationship
hypothesis is not in itself sufficient for the prediction of health
risk. It is also necessary to assume that all members of the exposed
population have equal sensitivity to the radiation insult so that the
expression of health risk is independent of how individual exposures
are distributed. This requirement is not wholly satisfied. As
documented in the BEIR report, differences in sensitivity do exist;
for example, children are more radiosensitive than adults. There are
two considerations, however, which help validate the application of
available mortality data to a consideration of health effects from
-------
C-14
fuel reprocessing. Some of these data (those taken from Hiroshima and
Nagasaki) reflect, to a limited extent, exposure of a relatively
heterogeneous population. More Importantly, even though the number of
health effects will be dependendent on the exact makeup of the
populations at risk, the relative order of importance of the various
pathways of exposure will not be very sensitive to the population
characteristics near a given fuel reprocessing plant. Finally, it
should be pointed out that the health risk estimates made here assume
that the expected radiation effects are independent of other environ-
mental stresses, which may be either unique to the population surrounding
fuel reprocessing facilities or unique to the exposed groups considered
in the BEIR report.
The numerical risk estimates used in this appendix are primarily
from the BEIR report. What must be emphasized is that though these
numbers may be used as the best available for the purpose of risk-cost
benefit analyses, they cannot be used to accurately predict the number
of casualities, for a given dose equivalent, the BEIR report estimates
i
a range for the health impact per million exposed persons. For example,
the BBIR results from a study of the major sources of cancer mortality
»3 g
data yield an absolute risk estimate of 54-123 deaths annually per 10
persons per rem for a 27-year followup period. Depending on the details
of the risk model used, the BEIR Committee's relative risk estimate is
~t
Absolute risk estimates are based on the reported number of cancer deaths
per rad that have been observed in exposed population groups, e.g.,
Hiroshima, Nagasaki, etc.
Relative risk estimates are baaed on the percentage increase of the
ambient cancer mortality per rem.
-------
C-15
160-450 deaths per 106 persons per rem. It is seen that the precision
of these estimates is at best about a factor of 3-4, even when applied
to sample populations studied on the basis of the same dose rates. The
application of the BEIR risk estimates to exposures at lower dose rates
and to population groups more heterogenous than those studied increases
the uncertainty in the risk estimates. Considering the limitations
of presently available data and the lack of an accepted theory of
radiocarcinogenesis, emphasis should be placed on the difference in
risk estimates between the various procedures and countermeasures
discussed in this report rather than on the absolute numbers. Where the
absolute numbers must be used for risk-cost-benefit balancing, it
should be remembered that these risk estimates are likely to be re-
vised as new information becomes available. Notwithstanding these
disclaimers, it is also pertinent to note that we are in a better
position to evaluate the true risks and the accompanying uncertainties
from low levels of radiation than from low concentrations of other
environmental pollutants which might affect populations in the vicinity
of a fuel reprocessing plant.
A. Dose-Risk Conversion Factors for Krypton-85
1. Total Body Dose-Risk
The BEIR report calculates the cancer mortality risk
(including leukemia mortality) from whole body radiation by two quite
different models. The absolute risk model predicts about 100 cancer
deaths per 10 person-rem while the relative risk model predicts between
160 and 450. An average cancer mortality of 300 annually per 106
person-rem would seem to be an appropriate mean for the relative risk
-------
C-16
model. The average of the absolute and relative risk models is 200,
which is close to the estimates of cancer mortality risk listed as
"most likely" by the Committee, Cancer mortality is not a measure of
the total cancer risk, which the Committee states is about twice that
of the yearly mortality.
The computation of health risk from continuous krypton-85
total body exposure used in this report is the one appropriate for
total body irradiation.
Estimated Cancer Risk from TotalBody Irradiation .
Cancer mortality = 200 deaths per year for 106 person-rem
annual exposure. Total cancers = 400 cancers per year for
10 person-rem annual exposure.
2. Gonadal Dose-Risk
The range of the risk estimates for genetic effects set forth
in the BEIR report is so large that such risks are better considered on a
relative basis for different exposure situations than in terms of
absolute numbers. The range of uncertainty for the "doubling dose"
(the dose required to double the natural mutation rate) is 10-fold
(from 20 to 200 rad); and because of the additional uncertainties in
(1) the fraction of presently observed genetic effects due to back-
ground radiation, and (2) the fraction of deleterious mutations
eliminated per generation, the overall uncertainty is about a factor of
25. The total number of individuals showing very serious genetic
effects such as congenital anomalies, constitutional and degenerative
-------
C-17
diseases, etc., is estimated at somewhere between 1,800 and 45,000 per
generation per rad of continuous exposure; i.e., 60-1,500 per year if
a 30-year generation time is assumed. This level of effect will not be
reached until after several generations of exposure; the risk to the
first generation postexposure is about a factor of 5 less.
The authors of the BEIR report reject the notion of "genetic death"
as a measure of radiation risk. Their risk analysis is in terms of
early and delayed effects observed post parturn and not early abortion,
stillbirths or reduced fecundity. . Because of the seriousness of the
genetic effects considered here, e.g., mongolism, the emotional and
financial stress would be somewhat similar to death impact. Indeed,
many of the effects described are those which lead directly to infant
mortality (fetal mortality is excluded). For the purposes of this
report this class of genetic effects will be considered on the same
basis as mortality data,
Estimated Serious Genetic Risk
from ContinuousGonadal Irradiation
Total risk = 300 effects per year for 106 person-rem annual
exposure.
Less serious genetic effects have also been considered by the
BEIR Committee. These have been quantified under the categorgy "unspeci-
fied ill health." The Committee states that a continuous exposure of
one rem per year would lead to an increase in ill health of between
3 and 30 percent.
-------
C-18
3. Lung Dose-Risk
Due to the insufficient data for the younger age groups,
estimates of lung cancer mortality in the Bill report are only for
that fraction of the population of age 10 or more. For the risk esti- '
mate made below, it is assumed that the fractional abundance for lung
tumors is the same for those irradiated at less than 10 years of age
as it is for those over 105. On an absolute risk basis lung cancer mortality
is about 26 deaths per annum per 10 ^ persons irradiated continuously
at a dose rate of 1 rem per year. This is a minimum value. The BEIR
report states that .the absolute risk estimates may be too low because
observation times for exposed persons are still relatively short compared
to the long latency period for lung cancer. Furthermore, lung cancer
risks calculated on the basis of relative risk would be larger. For
the risk estimate made here, it is assumed that the ratio of absolute risk
to the average relative risk is at least a factor of 0.5, i.e., the
same ratio as in the case of total body irradiation discussed above.
Estimated Lung Cancer Risk '
fromContinuous Lung Irradiation
Lung cancer mortality = 50 deaths per year for 10 person-rem
annual exposure.
4. Skin Dose-Bisk
There is no doubt that the dose to the skin delivered by
krypton-85 exceeds by about two orders of magnitude the insult to other
organs. However, epidemiological evidence of any real risk from such
5An absolute risk estimate is not very sensitive to the inclusions of this
assumption since lung cancer incidence is very small in the young.
-------
• ; C-19
insults at the dose levels ebnsidered here is nonexistent. This is
not to say that the linear dose-effect, assumption does not,hold for
skin cancer but rather that the.BEIR Committee found that from the
extensive evidence, they examined "numerical.estimates of risk at low
dose levels would not seem to be warranted." The authors of this report
concurred with the BEIR report. However, rather than defining a
zero risk per rad for any radiation insult from krypton-85, an upper
limit of risk is proposed. It should be noted that skin cancers are
rarely fatal and usually not very debilitating. The estimated
upper risk for continuous exposure is:
C C
Skin cancer - upper limit = 3 skin cancers" per year for 10° perspn-
rem annual exposure.
B. Dose-Risk Conversion Factors for Tritium
1. Total Body Dose-Risk :
The somatic dose-effects from tritium are not expected to be
unique. Risk estimates for total body irradiation are based on the
same information reviewed in Sectioii. A for krypton-85 total body
exposure.
2. Gonadal Dose-Risk
The genetic risk from tritium per unit gonadal dose is expected
to be the same as for the beta and gamma radiation from other isotopes.
Some experiments with bacteria (Ref. C-24) have shown that the location
of a tritium atom on a particular DNA base can enhance the mutation rate,
5Assuming 30 years at risk exposure and that the incidence of skin cancer
will be 10% of all radiation-produced cancers except leukemia, breast,
lung, G.I. tract, and bone cancers.
-------
C-20
However, if it is assumed that tritium labeling is a random phenomenon,
the percentage for such locations that are specifically labeled will be
extremely small at the exposure levels considered here. Therefore, the
risk estimates for gonadal .irradiation from krypton-85 listed in section
'A are also appropriate for estimating the genetic risk from tritium .
exposures,
C. Dose-Risk Conversion Factors for lQdine-129 andIodine-131
Iodine is concentrated in the human thyroid. Therefore, the insult
from radioiodines is important only for the thyroid. The dose to other
organs is orders of magnitude less. Two health effects follow high
level exposures of thyroid tissue to ionizing radiation: benign neoplasms
and thyroid cancer. Though the former are a more common radiation
effect, only the risk from cancer is considered here.
While children are particularly sensitive to radiation damage to
their thyroid glands, thyroid cancer is not usually a deadly disease
for persons in younger age groups but mortality approaches 25% in
persons well past middle age. It is not presently known if the radi-
ation-induced cancers which are more frequent for persons irradiated
early in life will follow the same patterns of late mortality.
The BEIR report provides risk estimates only for morbidity (not
mortality) and only for persons under 9 years of agej i.e., 1.6-
9.3 cancers per 10^ person-rem years. From the Hiroshima data and
other studies it would appear that for persons over 20 years old the
radiation-induced thyroid cancer incidence is much lower and may
approach zero.
-------
C-21
Since information in the BEIR report Is not sufficient in itself
to estimate the cancer incidence from continuous exposure, tentative
risk estimates for this study are also based on risk estimates described
in ICRP'Publication No. '8 (Ref. C-25) as well as the mean of the BEIR
Committee's various estimates of incidence per rem. Infants and fetuses,
composing,approximately 2.5 percent of the population, are, of course, the
most sensitive group. For this group about 150 thyroid cancers may accrue
annually per 10 persdn-rem annual exposure. For the approximately
401 of the populations that is in the 1-19 year age group it is assumed
that the incidence is a factor of about 4 less, and that for the balance
of the population, it is a factor of 30 less,
EstimatedThyroid Cancer Risk from Radioiodines
Morbidity for less than age 1:
150 cases per year for 10 person-rem annual exposure.
Morbidity for less than age 20:
35 cases per year for 10 person-rem annual exposure.
Morbidity for more than age 20:
, 5 cases per year for 106 person-rem annual exposure.
It is unlikely that the annual mortality from this cancer would be
much larger than 25% of the rate of incidence. As for other radiation
effects, a true measure of the risk from thyroid cancers would-be life
shortening, but insufficient mortality data prevents such an approach.
D. Dose-Risk Conversion Factors for Plutonium and other:Actinides
The lung cancer mortality risk discussed in Section A is the best
available information on the consequences of lung exposure. While it
is based on mortality information from miners exposed to alpha emissions
from particulates as well as more conventional dose-risk data, it is
-------
C-22 .
probably not really adequate for describing the risk from inhaled
plutonium. There is good evidence that a fraction of such particulates
are cleared from the lungs and relocated into the respiratory lymph
nodes. The organ dose received by a lymph node in this case is not
really known, but is probably on the order of 50 times the average
dose to lung tissue. The ICKP does not consider these highly irradiated
nodes to be the organ at maximum risk and preliminary results of animal
experiments would tend to confirm their judgment. However, it is not
a settled question.
Even if the lung is the critical organ for such exposures, there
is little reason to believe that the average lung dose, presently used
in health-risk analyses, is really relevant to estimating the risk from
air-borne particles. Estimates of the actual dose from discrete sources
of alpha radiation are subject to large variability simply because
little is known about the volume over which the energy deposition takes
place. Even though as many as 4x10 particles (0.2 Pm diameter) of
plutonium-239 are required to deliver an average energy deposition of
1 rad to the lung, the dose is not evenly distributed; only about 0.2%
of the lung volume absorbs the emitted energy. Health risk estimates
based on dosimetry are probably unwarranted under these circumstances
and use of a body burden approach to health-risk assessments would
appear to be a more likely route to Success, Unfortunately, experiments
allowing this approach are not yet complete. Therefore, the lung
cancer risk estimate for exposure to actinides for purposes of this
study is as given in Section A for uniform lung exposure.
-------
C-23
The dose conversion factor for a soluble form of plutonium will
differ from that presented for insoluble plutonium for a given air
concentration. However, the associated risk (expressed in effects
induced per unit air concentration) resulting from the soluble form
is expected to be the same order as for the insoluble form as
analyzed here.
-------
C-24
REFERENCES
C-l U.S. Atomic Energy Commission, Meteorologyand Atomic Energy - 1968;
Slade, D.H., Editor, Division of Reactor Development and Technology,
July, 1968.
G-2 Turner, D. Bruce (ESSA), Workbook, of Atmospheric Dispersion Estimates,
U.S. Department of Health," Education, and Welfare, Public Health
Service, Consumer Protection and Environmental Health Service,
National Air Pollution Control Administration, Cincinatti,
Revised 1969.
C-3 Garner, R.J., and Russell, R.S., Isotopes of Iodine,
Radioactivity and HumanDiet, ed. R.S. Russell, Pergamon Press, 1966.
C-4 Bryant, P.M., Data for Assessments Concerning Controlled and Accidental
Releases of i31I and 137Cs to Atmosphere, Health Physics, Vol. 17,
pp. 51-57, 1969..
C-5 Bryant, P.M., Derivation of Working Limits for Continuous Release
Rate of Iodine-131 to Atmosphere in Milk Producing Area,
Health Physics, Vol. 10, pp. 249-257, 1964.
C-6 Bryant, P.M., Derivation of Working Limits for Continuous Release
Rates of 129I to Atmosphere, HealthPhysics, Vol. 19
pp. 61.1-616, 1970.
C-7 Federal Radiation Council, Background Material for the Development
of Radiation Protection Standards, Staff Report No. 1, May 13, 1960.
C-8 Kirk, W. P., Krypton-85 - A Review of the Literature and an Analysis
of Radiation Hazards, Environmental Protection Agency, Office of
Research and Monitoring, Eastern Environmental Radiation Laboratory,
January, 1972.
C-9 International Commission on Radiological Protection (ICRP), Recommenda-
tions of the International Commission on Radiological Protection, Report
of Committee II on Permissible Dose for Internal Radiation, ICRP Publica-
tion 2, Pergamon Press, 1959.
C-10 Evans, A.G., New Dose Estimates from Chronic Tritium Exposures,
Health Physics. Vol. 16, pp. 57-63, 1969.
C-ll Koranda, J.J., and Martin, .R., Persistence of Radionuclides at Sites
of Nuclear Detonations, BiologicalImplications of the Nuclear Age,
U.S. Atomic Energy Commission Symposium Series No. 5, 1965.
-------
C-25
C-12 Bond, V.P., and Feinendegen, L.E., Intranuclear % Thymidine, Dosimetrie,
Radiobiological and Radiation Protection Aspects, Health Physics, Vol. 12
pp. 1007-1020, 1966.
C-13 Tompkins, P.C., Environmental Radiation Protection Criteria and
Tritium, Paper presented at Tritium Symposium, Las Vegas, Nevada,
August, 1971.
C-14 National Council on Radiation Protection and Measurements, Basic
Radiation Protection Criteria,"CEP Report No. 39, January, 1971.,
C-15 Federal Radiation Council, Background Material for the Development
of Radiation Protection Standards, Staff Report No. 2, May 13, 1960.
C-16 Federal Radiation Council, Background Material for the Development
of Radiation Protection Standards, Staff Report No. 5, July, 1964.
C-17 Durbin, P.W., Lynch, J., and Murray, S., Average Milk and Mineral
Intakes (Calcium, Phosphates, Sodium and Potassium) of Infants in the
United States from 1954-1968: Implications for Estimating Annual
Intake of Radionuclides, Health Physics, Vol. 19, pp. 187-222, 1970.
C-18 McClendon, J.F., Iodine and the Incidence ofGoiter, University of
Minnesota Press, Minneapolis, 1939.
C-19 Bair, W.J., Plutonium Inhalation Studies, Battelle Northwest Laboratory,
BNWL 1221, 1970.
C-20 'International Commission on Radiological Protection, Publication No.6
Pergamon Press, 1964,
C-21 Dolphin, G.W., The Biological Problems in the Radiological Protection
of Workers Exposed to 23T>u, Health Physics, Vol. 20
pp. 549-557, 1971.
C-22 Snyder, W., Internal Exposure, Chapter 10, Principles of Radiation Pro-
tection, K.Z. Morgan & J. Turner, John Wiley and Sons, 1967.
C-23 National Academy of Sciences - National Research Council, The Effects
on Populations of Exposure to Low Levels of Ionizing Radiation,
Report of the Advisory Committee on the Biological Effects of Ionizing
Radiation (BEIR), U.S. Government Printing Office 1972.
C-24 Funk, F., Cytosine to Thymine Transitions from Decay of Cytosine-53H
in Bacteriophage S 13, Science, Vol. 166, pp. 1629-1631, 1969.
C-25 International Committee on Radiological Protection, Publication No.8
Pergamon Press.
-------
-------
APPENDIX D
Radiation and Economic Impact
Analysis Techniques
-------
-------
B-l
I. INTRODUCTION
This appendix presents the calculations! techniques uspd in
arriving at radiation dose and health risk estimates for the general
population due to nuclear fuel reprocessing and the general approach
taken to assess the economic impact of effluent controls on the
industry. The dose and risk estimates are made for individuals
living at 3 kilometers from a reprocessing plant, the regional popu-
lation within 80 kilometers (50 miles) of a reprocessing plant, for
the total United States population (including that of the regional
zone), and for the world population (excluding the U.S. population),
The dose and risk estimates are made for a representative single five
tonne per day plant and for the projected industry as a whole. Only
releases via the atmospheric route are considered since future plants
are not expected to release significant quantities of radioactive
material in liquid effluents. One present facility has significant
liquid releases and this case should be considered on an individual
basis.
-------
D-2
II. ESTIMATED REGIONAL POPULATION EFFECTS
The regional population around a reprocessing plant may be
exposed to higher radioactive material concentrations due to releases
from the plant than is the total U.S. population. Thus, this
regional population group is considered as a special case. The total
regional population dose received by a specific body organ (organ i)
exposure to a specific radionuclide (radionuclide j) is estimated by
using the following equation:
D = 0 (20 F. C ?„ person-rem/yr
i i B.
.
(3 km) ij
Where :
Q.= Release rate of radionuclide j from the fuel reprocessing plant
in curies released per second (Ci/s) . . The average annual
release rate is determined by using the curie/tonne values
in appendix A, the plant decontamination factors in appendix
B, and the size of the plant in tonne/day . capacity . For this
evaluation, the capacity of a single plant is taken as
5 tonne/day which is equivalent to 1500 tonne/year. . It
is assumed that a plant operates at full capacity.
= The meteorological dispersion factor, i.e., ttie radionuclide
(3 km)
concentration in air, X, 3 km from the plant per unit of
3
release rate from the plant [(yCi/cm ) /(.Ci released/s) ] .
A distance of 3 km is chosen as a reference point since the
air concentration at this point is generally not significantly
affected by the stack height of the plant. Using the methods
referenced in appendix C, X/Q values have been calculated for
-------
D-3
many sites by many people. In this evaluation, a value for
(X/Q)3km) of 5 X 10~8 ( Ci/cm3)/(Ci/s) is considered to be
representative of a typical plant. This value is assumed to
be uniform in all directions around the plant. Actual plant
values may vary from the above value by as much as factors of
five or ten.
F » The pathway dependent dose conversion factor which gives the
dose to organ i due to a medium (air, food, or water) concen-
2
tration of radionuclide j [(rem/year)/( Ci/cm )]. For
example, iodine exposure of the thyroid gland by inhalation
and milk ingestion would have two separate dose conversion
factors. The dose conversion factors used in this study are
given in appendix C. For radionuclides with long retention
times in the body, e.g., the actinides, the conversion factor
represents the equilibrium dose rate resulting from a continu-
ous factor represents the equilibrium dose rate resulting
from a continuous constant intake for several years.
C « The regional dilution and population distribution correction
factor. It is a ratio of the average individual dose within
80 km of the plant to the average individual dose at 3 km. It
takes into account increased dilution as the radionuclides are
transported further from the plant and an uneven distribution
of population around the plant. It can be calculated
theoretically by assuming a population distribution (lef,
D-l) or it can be determined from population dose calculations
around real plant sites, for this study, the results of dose
-------
D-4
calculations for about 50 reactor.sites (Ref« D-2) were
analyzed and a value of 0.028 (rem/person within 80 km)/
rem/person at 3 km) was obtained. Theoretical results are
similar. This value was assumed to be the same for all
nuclides. Fuel reprocessing plants are assumed to be located
at sites with population characteristics similar to those at
reactor sites. Individual plant correction factors may vary
by as much as a factor of five from the average value given
above. The distance of 80 km (50 miles) was chosen as a cut-
off on regional calculations since the distance is large enough
to include any nearby large population center yet small enough
so that the area affected can be considered a local area.
R = The populations within 80 km of the fuel reprocessing plant
site. The population values of the above-mentioned 50 reactor
sites, taken primarily from environmental reports, lead to an
average population around a site of 1.5 x 10 people in 1980.
Population sizes around individual plants can vary from this
by a factor of three. The doubling time of this population
is assumed to be about 40 years. For purposes of characterizing
age-specific factors, 2.5% of the population is considered to be
under 1 year old, 45% between 1 and 20 years old, and the remainder
over 20. One-half of the population is used for genetically signi-
ficant" dose calculations because of the child-bearing potential.
Using the'above factors, the dose equation becomes:
-------
D-5
D.. = (Q Ci/s) x
J J
x (F.. ^7%) x (0.028) x (1.5 x 106 people)
ij yCi/cm-3
x (1.5 to take into account population growth)
= 0.0032 Q. F person-rem/year
This represents the average annual population dose for a 40
year period of constant emissions during constant plant operation.
The equation can be modified further by considering that:
A1 x 1500
3.15 x 10? x d
A1 x 1500 Ci/s
,
where:
Aj = Curies of radionuclide j per tonne of fuel reprocessed
(appendix A)
dj = Plant system decontamination factor for radionuclide j
(appendix B) and 1500 tonne are processed per year at a 5
tonne per day capacity, and there are 3.17 x 10 seconds per year)
Therefore:
-7-L
^li ~ •'••-' x ^-0
-------
D-6
where the J. values are those given in appendix C. The total
regional health risk for one plant, H, is:
Hi = (Ji V
The total annual fuel reprocessing industry effect on the
regional populations is obtained by multiplying the effect for
one plant by the number of operating plants in any given year
(appendix A).
-------
D-7
III. ESTIMATED EFFECTS OK THE UNITED ' STATES POPULATION
Several radionuclid.es released in the gaseous effluents from
a fuel reprocessing- plant spread from the local region to all or
part of the total United States land area before being diluted
around the world, . The method of estimating effects to the U.S.
population depends on the radionuclide being, considered. The radio-
nuclides considered in this study are tritium, krypton-85, iodine-129,
and the actinides. Iodine-131 is considered to be only a regional
problem because of its short half-life and the rate of local deposi-
tion assumed. Krypton-85 and tritium expose 'the U.S. population
only briefly before subsequent dispersion throughout the world, while
iodine-129 and the actinides are assumed to build up only on U.S. soil.
In fact, the impact of these latter two may be restricted to a region.
A. Tritium
It is assumed that most of the tritium released as a gaseous
effluent from the reprocessing plant will enter into the hydrologic
cycle. The released tritium is assumed to fall out or rain out over
the Eastern United States [3.7 x 10 6 km2 (1.5 x 10 6 mi )] and becomes
diluted by the annual rainfall [40 inches (average 100 cm) ] (Ref>
D-3) over this area. With some further dilution by uncontaminated
water, this then becomes the water concentration to which the popula-
tion of the Eastern United States (80% of the total U.S. population)
along with their animals and crops are exposed.
The total population dose is then given by:
D.. = (C. ) x f x (F.. ) x 0.80 P person-rem/yr
13 3 ml ij pCi/ml u
-------
D-8
where:
C, - The water concentration of tritium determined by diluting the
yearly input, Q. (Ci/s) x 3.15 x 10 s/yr, of tritium to
the environment by the average annual rainfall over one-half
of the United States.
I - A factor to take into account dilution of tritium by uneontamin-
ated water from deep artesian wells and the fact that not all
tritium will fall out over the Eastern United States. Some,
perhaps a large portion, will fallout over the Atlantic Ocean
and will be diluted in a larger volume. For this evaluation
f = 0.5 is used although it is only an estimate.
.. = Dose conversion factor from appendix G [100 (rem/yr)/(uCi/ml)
for this case.]
= Population of the United States. The population growth of the
U.S. is estimated as in figure D.I. One-half P is used for
genetically significant dose calculations.
Tritium concentration, C., is related to the environmental
input by;
3.15 x 107 s/yr
(1.5x10° miz)x(1.61x10^ cm/mi)%(40 in)x(2.54 cm/in)
8,0 x 10~b Q pCi/ml
2,5 x 10~13 -ji N pCi/ttl
-------
D-9
Z
o
Cu
O
Q_
CO
oo
Q
LLJ
fc=
Z
D
Q
LU
I—
u
LU
O
Of.
Q_
3
O)
NOIXVindOd S3J.V1S QBIINH
-------
D-10
where:
j = Curies/tonne - froto appendix A.
D. = System decontamination factor from appendix B.
N = Number of tonnes processed per year. (1500 tonne/yr for a
single plant and as in appendix A for the total industry.)
Therefore:
'A. person-rem/yr to U. S. A.
D,, = 2.5 x 10 x 0.80 f N — F., P population from the
13 d. 1J U fallout of tritium
"B. Rrypton-85
Part of the population of the Eastern United States is exposed
to air concentrations of krypton-85. as it passes from the fuel
reprocessing-plant to'the Atlantic Ocean on its first pass around
the world. The dose from this exposure pathway is taken from a
study recently performed at the National Oceanic and Atmospheric
Administration (Reference D-4). For a plant in Morris, Illinois,
releasing one curie of'krypton-85 'per year, the" population-weighted
concentration on its first pass over the Eastern United States to
-16 3
the Atlantic Ocean is"2.5 x 10 person-Ci/cm . For purposes
of this evaluation,•this value is considered adequate to use for
all plants. This value is then'multiplied by total annual releases
and dose conversion 'factors to obtain dose values-.
-------
B-ll
person-_Ci
±j -
where N, A , d . , and F. . are as defined previously. The 1.5 factor
«J J J
is to account for population growth based on a doubling time of 40
years .
Therefore :
~10
A.
D. . = 3.8 x 10 •"* N -L F. .
ijj d. ij
person-rem/yr to U. S,
population from release
- 1 of krypton-85
C. Iodine-129
As a first approximation, all of the iodine-129 release is
/- *\
assumed to fall out over the Eastern United States [3.75 x 10 km
fi O
(1.5 x 10 mi) ] and uniformly mix with the stable iodine in the
soil to a depth of 20 cm. This then becomes the specific activity
of iodine in the diet to which all persons in this part of the
country are exposed.i Because of its long half-life (1.6 x 10 years),
iodine-129 will build up on the soil and expose the population long
after it has been released to the environment. The movement of
iodine-129 in the biosphere is not well documented at the present
time. However, even with the assumptions that must be made, the
first estimate obtained is considered reasonable.
-------
B-12
The specific activity of iodine-129 in the soil at time t,
, 129T/ 127T ,
curies I/gr I, is:
Specific _ Z H(f) ' Aj *; f
-
Activity - o d. ' l>5 x 1Qb ^ x (]U(j x 1()5
4 x 1°~6 S 127I » 1.5 g soil on , ^
a - — • •« — - — x 20 cm depth
g soil .3 r
° cm
- 2.2 x 10~13 I N(f ) AJ f Ci 129I
o d. g 127i
where £ is the fraction of iodine-129 release that stays in the
soil, and it also includes dilution by iodine taken in from other
sources, and the other factors are as defined previously. For
purposes of this evaluation, f .is taken as 0.5. This value is only
a rough estimate, however, and may be considerably in error. Like-
—6 127
wise, the value of 4 x 10 g I/g soil is taken from only one
Ref . (D-5) and may not be representative of the Eastern United
States, These numbers can easily be changed when better values
are determined. Using the dose conversion factor given in
appendix C, the annual thyroid population dose rate as a
function of time, t, is determined from:
r,129
D .- (specific activity %r-^ x (4.4x10^ .... )
ij g127! Ci129I/g127I
x (0.80 P (t))
-8 i
= 7.7 x 10 P (t) Z I(f*) —*• f thyroid person-rem/yr
-------
D-13
This is the annual population dose (thyroid) due to the total
amount of iodine-129 in the environment at the time of calculation.
For purposes of determining health risk, the iodine-129 is assumed
to expose the population for 100 years beyond the time of release.
D. Actinides
, The actinides are assumed to build up in the Eastern United
States in the same manner as iodine-129 with ..the principal ex-
posure pathway believed to be resuspension of the material and
inhalation. The annual population dose to the lungs from the
buildup of actinides in the environment is;
E 0.8 P (t) N(t') l f R F
t'»o U d, 8
D . 3
J 3.8 x 10b km2
3 2
where R is the resuspension factor in.terms of Ci/m air per Ci/m
5
on the ground. Based on calculations using fallout data and data
— H
from around Rocky Flats (Reference D-6), a value of R - 1Q is
S
used in this study. The uncertainity.of this value may be of the
order of several orders of magnitude. The fraction of actinides
released that remains on the soil is taken as 0.5 for this study.
Therefore:
D. ,
-19 t"
= 2.1 x 10 P (t)f I N(t')
0
A F
2> 3 .ij
J dj
person-rem
year
-------
D-14
The dose calculation is performed for plutonium-239 and the total
of all the long-lived actinides is about 10 times the plutonium-239
dose for long-term exposure. This results from a consideration of A.
J
and F.. values in appendixes A and C, For purposes of determining
health risk, the actinides are assumed to expose the population for
100 years beyond the time of release.
The health risks from all of these pathways are estimated by:
>.. J.
Admittedly, the pathway models used in this section for the
determination of environmental effects to the United States popu-
lation are uncertain and unproven. They are presented as a first
order approximation of these effects and the points of uncertainty,
especially relative to environmental behavior, are indicative of
research needs already identified in these areas.
-------
D-15
IV. EFFECTS ON THE WORLD POPULATION OUTSIDE THE U.S.
Two radionuclides released in fuel reprocessing are expected to
result in doses to the world population - krypton-85 and tritium.
A. Krypton-85
The worldwide dose due to krypton-85 exposure can be estimated
by diluting the output froift one year of fuel reprocessing into the
world's atmosphere (5.14 x 10 g; sea level air density » 0.00129
o
g/cm ) (Ref. D-7), and then determining population dose while it
decays away. The total yearly dose rate at any instant in time is
a combination of contributions from all previous environmental
releases of krypton-85. The total population dose to be received
by a specific body organ (organ i) over the total time of decay from
a one year's input of krypton-85 is given by;
t, AJ % / ^ ir,6 PCi s .0.00129 g/cm3 rem/yr
Dljr (decay) - <— x 10 ^= ) x (5<14 x ^L g) * (F. C1/m3) x
3
/ Pw(t) x e~xt'dt'"
where:
X - Decay constant for krypton-85 (0.0645/yr)
P (t)' - World population at time t; (t-0 at time of release). The
9
world population is taken as 3.56 x 10 people in 1970 with
an annual growth rate of 1.9%/yr (Ref. D-8). For genetically
significant dose calculations, one-half the world population
is used. Five percent of the world effect is subtracted to
account for the United States contribution to world population dose,
The other factors are as defined previously.
-------
D-16
Therefore:
D£. (decay) = 1.9 x l(
8
. e
°*019t
tdtal wbrld persott~fem
contaitted by one year's
release (less U^ S, dose),
where t" is measured from 1970 to the time of release.
The annual dose rate at time t from the buildup of krypton-85
in the atmosphere from releases i's given bys
D (buildup) =
. 1A6 -0.0645(t-f)
j x 10 x e s '
o d.
x
[
O.OQ129
5.14 x 1021 g
ij w
where t=t'=0 in 1970.
Thus: D.. (buildup) - 8.5 x lO'10 - F. ,-°-0455tx Z N(tOe*°*°645t'
person-rem/yt to the world's population (less U
-------
B-17
19
the Northern Hemisphere (one-half of 2.7 x 10 1) (Ref, D-7)
and assuming that the Northern Hemisphere's population (80% of the
world population) (Ref. D-8) is exposed to the resulting
concentration. The U.S. contribution (about 7%) is subtracted out.
The total population dose for organ i that will be received
over the total time of decay from a one year's input of tritium
is given by:
NA. 6 ci 2
D . (decay) - (—^ x 10 ^~) x ( — )
• 1J j y 2.7 x 10xy x 10J ml
r „ ,.,, -0.0562t* *
NA
D, .(decay) '- 5.3 x 10"6 -J- F.. e0'019t
Q
.
world person-rem
committed
by one year's release of tritium (less U. S. dose) where t is
measured from 1970 to the time of release.
The annual dose -rate at time t from the buildup of U. S. nuclear
power generated tritium released to the environment is given by:
n A. -1J N j H -Q.0562(t-t")
D.. (buildup) =* Z —»-: J- x 10 e
13
x f 2 > x F x P (t)
V* ntj J "f -| T£r
2.7 x 1022 ml 3
where: t = t» = 0 in 1970,
-------
D-18
DI. (buildup) - 0.20 x Hf6 f- I e-°'0372t Z N(t') e40'0562'"
j J o '
person-rem/yr to world's population from U. §. reactor produced
tritium (less U. S. dose) . For genetically significant doses use
one-half this value.
Again,, to evaluate future doses due to the environmental
-0.0562T 0,0191
buildup of tritium, multiply .by e '. e and integrate
from T==o to T-«> to account for population growth and decay.
; The health effects, resulting, from exposure to tritium and
krypton-85 are determined by:
H = V" D • . J*
2- ^ i
(Some radionuclide fuel content and system decontamination
factors used in these calculations are given in table D,l)
-------
D-19
Table D.I
Factors Used in the Assessment of Environmental
Radiological Impact of Fuel Reprocessing
H-3
Kr-85
1-129
1-131
Actinides
Activity
in Fuel
(Ci/MT)
800.00
10,500.00
0.04
2.00
(See Appendix A
Plant
Decontamination
Factor
1.0
1.0
1000
1000
•••• io9 •
fable A.4)
-------
D-20
V. ECONOMIC ANALYSES
Annual-costs should include considerations of debt service,
depreciation, and Federal taxes. The first twd are dependent on
the assumed plant life, salvage value, investment capital and
debt. An example of the total annual costs for a 5 tonne/day
plant is presented in table D-2 (Ref. D-9). In this regard, the
total economic impact upon the investors of the inclusion of
a particular control system requires consideration of the cost of
money invested over the life of the plant. This, of course, will
be somewhat offset by the increased value of the product over
the same period.
Several methods are available for presenting the cost ascrib-
able solely to control systems. These methods are described as
annual cost, present worth, and total commitment.
In order to obtain estimates of these values, it is necessary
to specify first costs (P), interest rate (i), debt lifetime (n)
plant life and estimated salvage value (L). In addition a discount
factor of 7.5% was used to estimate present worth. For the present
analysis, the lifetime is considered to be 40 years, the salvage value
is considered negligible, and the effective interest rate is assumed to
be either 10% or 24%. These estimates of costs are based upon the
assumption that the control systems are add-on facilities and not directly
required for processing.
The annual costs are considered to be the sum of capital
recovery costs and the annual operating expenses (A.O.E.). Annual
costs therefore are estimated from the following equation:
Annual cost = P x erf -f A.O.I,
-------
TABLE D.2
Estimate of the Economics of a 5 MTU/Day Reprocessing Plant—7
(Note that "000" is omitted from all
Tear '
1971 -2
1972 -1
1973 0
1974 1
1975 2
1976 3
1977 4
1978 5
1979 6
1980 7
1981 8
1982 9
1983 10' •
1984 11
1985 12
1986 13
1987 14
1988 15
Fuel
Processed,
Metric
Tons /Year
290
690
1,200
1,320
1,500
1,500
1,500
, 1,500
1,500
1,500
- '1,500
1..500
: 1,500
1,500
1,500
Operating
Costs
7,708
9,975
12,070
12,520
13,215
13,215
13,215
13,215
13,215
13,215
13,215
13,215
13,215
13,215-
13,215
Debt
Service
2,567
2,567
2,567
2,567
2,567
2,567
2,567
2,567
2,567
2,567 -
2,567
2,567
2,567
2,567
2,567
Depre- ,
elation—
9,750
9,130
8,420
7,800
7,170
6,500
5,840
5,220
4,550 '
3,500
3,270
2,570
1,950"
1,330
620
Other
Fixed ,
Co-sts5-'
2,430
2,430
2,430
2,430
2,430
2,430
2,430
2,430
2,430
2,430 '
2,430
2,430
2,430
2,430
2,430
Total
Costs
22,455
24,102
25,487
25,317
25,382
24,712
24,052
23,432
22,762
22,112
21,482
20,782
20,162
19,542
18,832
money values)
Gross
Receipts
at
$27.50/kg
7,975
19,000
33,000
36,300
41,250
41,250
41,250
41,250
41,250
41,250
41,250
41,250
41,250
41,250
41,250
Before-
Tax Income
(14,480)
(5,120)
7,513
10,983
15,868
16,538
17,198
17,818
18,468
19,138
19,768
20,468
21,088
21,708
22,418
Federal
fax4'
7,090
7,940
8,250
8,560
8,860
9,200
9,480
9,820
10,100
10,400
10,780
Cash Flow
(18,200)
(18,200)
(18,200)
(4,730)
4,028
15,933
18,783
15,948
15,098
14,788
14,478
14,158
13,838
13,558
13,218
12,938
12,638
12,258
Obliga-
tion
-19,660
-42,470
-68,870
-85,370
-94,342
-90,939
-83,685
-78,568
-73,590
-68,202 o
-62,324 N>
-55,856 *"*
. -48,748
-40,820
-32,022
-22,134
-11,016
+1,242
— 70 percent equity at 16 percent; 30 percent debt at 7 percent.
— Sum-of-year digits.
c/
— State ^nd local taxes (2.51 percent),-'insurance (0.25 percent), and interim replacement (0.3,5 -percent).
— At 48 percent.
From - Lecture VIJ. - Nuclear Fuel Processing, G.G. Eichhole, 1972 Short Course at Georgia Tech., re: Nuclear Fuels-Management and Economics
-------
D-22
where P and A.O.E. are obtained from the estimates presented in table
B-2. The capital recovery factor (erf) for 40 years at 10% is equal
to .105 (Ref. D-10).
Annual costs can be converted to present worth of all
disbursements for lifetime, n. This is accomplished by multiplication
of the annual costs by the present worth factor (pwf) for the
appropriate lifetime and interest rate and by adding the result to the
first cost; PW = A.C. (pwf) + P.
The present worth factor for 40 years at 10% is 9.779 (Ref. D-10).
The total control system cost can also be estimated over the life
of the plant. Assuming that the principal cost is repaid within 10
years at an interest rate of 10%, the total cost over 40 years is
estimated to be as follows:
T.C. = (P x erf 1 ) x 10 + A.O.E. x 40
where: erf = .163
The system cost presented in table D-3 is in terms of total cost for
both 10% interest with 10 yr debt life and 24% interest with 20 yr debt
life. In addition, the present worth for the latter parameters is
presented. These figures do not consider the loss of opportunity
costs.
The costs to the investors in time and money associated with
litigation because of suits by opponents are not presently known, but
could be substantial. Introduction of expensive control systems
during the initial construction stage and early use might be cost-
effective in an overall view even if the system might not be cost-
effective on the basis of risk reduction alone.
-------
TABLE D.3
CONTROL SYSTEM COSTS
First Cost
10% Int.
10 yr. debt
40 yr. life
Total Cost
24% eff. int.
20 yr. debt
40 yr. life
Total Cost
241 eff. int.
20 yr. debt
7.5% discount
40 yr. life
..P. W.
Isotope
Krypton-85
Iodines
Actinides
System (10 Dollar)
•. Cryogenic
•'• Distillation - 3
' Cryogenic Adsorption 3
Freon-Adsorption 1.5
; Caustic Scrubber
and AgZ Adsorption 1.2
HEPA Filter (2) .1
HEPA + Sand .35
(10 Dollar)
8.9
11.1
6.4
6
2.2
'• 3.3
(10 Dollar)
19
21
11
10
2.5
4.5
(10 Dollar)
11.5
12.1
6.4
5.4
1
2.1
*
e
to
Tritium
Voloxidation
10
20
54
31
The present worth of a caustic scrubber alone is estimated to he $3*85 x 10 ,
-------
D-24
Beyond the normal costs considered, there is also the matter that
failure to introduce control systems for effluents that could have
worldwide distribution (tritium and krypton-85) could be costly to the
Federal government in terms of foreign policy expenditures, The
magnitude of these costs is unclear at present, however.
The impact of fuel reprocessing plants upon the economy, e.g., on
industries which depend upon the economy, e.g., on industries which
depend upon the use of radiosensitive materials has been estimated to
be negligible as long as the media concentration of radioisotopes are
lower than the constraining levels set for humans by ICRP (Ref. B-ll,
D-12),
Cost Effectiveness Considerations. The effectiveness of risk
reduction of the previously described isotope specific control
(appendiK B) is dependent upon: (1) the group at risk which is
considered, and (2) the specific health effects used as a risk index.
The rate of decrease in number of health effects avoided by sequential
addition of the individual effluent control systems is dependent upon
the orcfef in which the systems are incorporated. The number of health
effects avoided by incorporation of the individual systems in the
order of decreasing unit cost-effectiveness is presented as a function
of the cumulative system control cost in figures D-2 and D-3. The
individual component for control of particulates, iodines, krypton,
3 3 •
and tritium were assumed to have decontamination factors 10 » 10 ,
2 2
10 , and 10 with present worth costs of 1, 5.4, 6.4, and 31 million
dollars respectively. A system capable of removing the residual
pollutants was assumed to be available with a cost of 20 million
-------
D-25
dollars also. Figure D-2 displays the reduction of the projected
risks of mortality (cancer induction and genetically coupled) as a
function of cumulative system costs for the regional, Eastern United
States, and worldwide population groups. The health risks considered
and presented in figure D-3 include nonfatal cancer induction as well.
The figures show that in every instance, the particulate control
system is the most cost-effective. The iodine control system is the
next most cost~effective, except for the situation which considers the
estimated mortalities to the world population which are avoided. When
considering the world population group, the iodine and krypton control
systems appear to be practically equally cost-effective for risk
reduction. In consideration of the regional and United States
population groups, tritium control appears the next most cost-
effective to the particulate and iodine systems although krypton
control is not significantly different from that of the tritium
control system.
The cost-effectiveness of a combined system for any group at risk
can be estimated from the slope of a directed line segment which
connects the origin and the appropriate arrow tip representing the
combination. The steeper the slope, the greater the cost-
effectiveness. It can be seen that the cost-effective excess of a
combined system will be necessarily less than that of its most cost-
effective component,
-------
D-26
a
LU
Q
O
t/}
t—
u
UJ
u_
u.
LU
1400-
1200-
1000-
800-
600-
400-
200
0
FIGURE D - 2 RJSK REDUCTiON(MORTALITY AND MORBIDITY)
vs CUMULATIVE CONTROL SYSTEM COST
NUCLEAR FUEL REPROCESSING - 40 YEAR OPERATION
5 TONNES/DAY THROUGHPUT
COMPONENT CONTROL SYSTEMS
P - PARTICULATE T - TRITIUM
I-IODINE R-RESIDUAL
K - KRYPTON
Total World Population
U.S. Population (excluding regional)
_ R
K
Regional Population (only)
(50 miles)
COMPONENT CONTROL SYSTEMS
ADDED IN THE ORDER OF
DECREASING COST EFFECTIVENESS
VP
o
1
10
1
20
i
30
i
40
I
50
I
60 7(
CUMULATIVE CONTROL SYSTEM COST
: (106 Dollars)
-------
B-27
FIGURE D - 3 RISK REDUCTION (MORTALITY) vs CUMULATIVE CONTROL SYSTEM COST
LU
P
Og
UJ
800-
600-
400-
200-
0.
NUCLEAR FUEL REPROCESSING - 40 YEAR OPERATION
5 TONNES/DAY THROUGHPUT
COMPONENT CONTROL SYSTEMS
P - PARTICULATE T - TRITIUM
I-IODINE R-RESIDUAL
K - KRYPTON
COMPONENT CONTROL SYSTEMS ADDED IN THE OF
DECREASING COST EFFECTIVENESS
Total World Population
K
U.S. Population (excluding regional)
i*"
R-
Regional Populational(only)
(50 miles)
0
10 20 30 40 50 60
CUMULATIVE CONTROL SYSTEM COST
(106 Dollars)
70
-------
D-28
REFERENCES
D-l Knox, J.B. Airborne Radiation from the Nuclear Power Industry,
Nuclear News, 14: 27-32, February 1971.
D-2 Gamertsfelder, G.C. Statement on the Selection of As Low As
Practical Design Objectives and Technical Speciations for
the Operation of Light Water Cooled Nuclear Power Reactors,
Presented at AEC Hearings on the "As Low As Practical Concept, "
1972.
D-3 Statistical Abstract of the United States, 1969,.U.S. Department
of Commerce.
D-4 Machta, L. National Oceanic and Atmospheric Administration,
Unpublished data.
D-5 Wayne, I,J., Demetrios A. Kontros, and W.D. Alexander. Clinical
Aspects of Iodine Metabolism, F.A. Davis Co., Philadelphia, Pa.
1964.
D-6 Compendium of Environmental Surveillance Around the Rocky Flats
Plutonium Plant, report prepared by FOD/ORP/EPA, November 1972.
D-7 Klement, A.W., Jr., C.R, Miller, R.P. Minx, and B. Sttleien.
Estimates of Ionizing Radiation Doses in the United States -
1960-2000, U.S. EPA document, CSD/ORP 72-1, 1972.
D-8 United Nations Statistical Office Report, 1966.
D-9 G.G. Eicholz. Lecture VII - Nuclear Fuel Reprocessing - 1972,
Short course re nuclear fuels, management and economics,
Georgia Tech.
D-10 Grant, E.I. and W. Ireson. Principles of Engineering Economy.
D-ll Frieser, H., G. Heiman, and E, Rane. The Effect of Radioactive
Nuclides of Photographic Emulsions, Photographische Korrespondenz,
98, 131-140, 1962.
B-12 Clear, Murray. Personal communication, Eastman Kodak Co.,
Rochester, New York.
AU,S, GOVERNMENT PRINTING OFFICE: 1973 546-311/111 1-3
-------
-------
-------
-------
-------
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
Washington, O.C. 20460
OFFICIAL BUSINESS
Return this sheet to above address, if you
do NOT wish to receive this material Q
or if change of address is needed Q (indi-
cate change, including ZIP code).
POSTAGE AND FEES PAID
U.S. ENVIRONMENTAL PROTECTION
AGENCY
EPA-335
THIRD CLASS BULK RATE
------- |